[ { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Construction zone\", \"B\": \"Service station\", \"C\": \"Urban\", \"D\": \"Highway\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000000", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Construction zone\", \"B\": \"Service station\", \"C\": \"Urban\", \"D\": \"Highway\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000000", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Low light\", \"C\": \"Partly cloudy\", \"D\": \"Foggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000001", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Low light\", \"C\": \"Partly cloudy\", \"D\": \"Foggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000001", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Light\", \"C\": \"Detour traffic\", \"D\": \"Heavy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000002", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Light\", \"C\": \"Detour traffic\", \"D\": \"Heavy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000002", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Sandstorm\", \"C\": \"Smoggy\", \"D\": \"Strong wind\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000003", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Sandstorm\", \"C\": \"Smoggy\", \"D\": \"Strong wind\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000003", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Countdown timer visible\", \"C\": \"Green\", \"D\": \"Dim / unclear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000004", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Countdown timer visible\", \"C\": \"Green\", \"D\": \"Dim / unclear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000004", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Uneven\", \"B\": \"Wet\", \"C\": \"Snowy\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000005", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000005.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000005.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Uneven\", \"B\": \"Wet\", \"C\": \"Snowy\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000005", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Bridge\", \"C\": \"Campus area\", \"D\": \"Service station\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000006", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Bridge\", \"C\": \"Campus area\", \"D\": \"Service station\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000006", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Partly cloudy\", \"B\": \"Dim streetlight visibility\", \"C\": \"Low light\", \"D\": \"Blinding sunlight\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000007", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Partly cloudy\", \"B\": \"Dim streetlight visibility\", \"C\": \"Low light\", \"D\": \"Blinding sunlight\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000007", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Lane merging\", \"C\": \"Vehicle queue\", \"D\": \"Accident blockage\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000008", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Lane merging\", \"C\": \"Vehicle queue\", \"D\": \"Accident blockage\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Moderate" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000008", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Mist\", \"B\": \"Hailing\", \"C\": \"Flurries\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000009", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Mist\", \"B\": \"Hailing\", \"C\": \"Flurries\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000009", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in this scene? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Off / malfunctioning\", \"C\": \"None\", \"D\": \"Temporary road sign light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000010", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in this scene? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Off / malfunctioning\", \"C\": \"None\", \"D\": \"Temporary road sign light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000010", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Wet\", \"B\": \"Dry\", \"C\": \"Icy\", \"D\": \"Potholes\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000011", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000013.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000013.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Wet\", \"B\": \"Dry\", \"C\": \"Icy\", \"D\": \"Potholes\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000011", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Highway\", \"B\": \"Roundabout\", \"C\": \"Forest road\", \"D\": \"Urban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000012", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Highway\", \"B\": \"Roundabout\", \"C\": \"Forest road\", \"D\": \"Urban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000012", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dusk\", \"B\": \"Blinding sunlight\", \"C\": \"Low light\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000013", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dusk\", \"B\": \"Blinding sunlight\", \"C\": \"Low light\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000013", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Moderate\", \"C\": \"Heavy\", \"D\": \"Lane merging\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000014", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Moderate\", \"C\": \"Heavy\", \"D\": \"Lane merging\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000014", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Overcast\", \"B\": \"Rainy\", \"C\": \"Flurries\", \"D\": \"Sleet\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000015", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Overcast\", \"B\": \"Rainy\", \"C\": \"Flurries\", \"D\": \"Sleet\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Rainy" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000015", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the frame? You should answer one choice from four answers \"A\": \"Pedestrian signal\", \"B\": \"None\", \"C\": \"Red\", \"D\": \"Covered by tree\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000016", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the frame? You should answer one choice from four answers \"A\": \"Pedestrian signal\", \"B\": \"None\", \"C\": \"Red\", \"D\": \"Covered by tree\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000016", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Construction debris\", \"C\": \"Gravel\", \"D\": \"Bumpy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000017", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000021.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000021.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Construction debris\", \"C\": \"Gravel\", \"D\": \"Bumpy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000017", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Construction zone\", \"C\": \"Rural\", \"D\": \"Airport roadway\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000018", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Construction zone\", \"C\": \"Rural\", \"D\": \"Airport roadway\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000018", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Backlit\", \"C\": \"Dusk\", \"D\": \"Foggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000019", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Backlit\", \"C\": \"Dusk\", \"D\": \"Foggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000019", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Rush hour\", \"B\": \"Moderate\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000020", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Rush hour\", \"B\": \"Moderate\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000020", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Smoggy\", \"B\": \"Strong wind\", \"C\": \"Sunny\", \"D\": \"Rainy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000021", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Smoggy\", \"B\": \"Strong wind\", \"C\": \"Sunny\", \"D\": \"Rainy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000021", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic light in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"None\", \"C\": \"Dim / unclear\", \"D\": \"Pedestrian signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000022", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic light in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"None\", \"C\": \"Dim / unclear\", \"D\": \"Pedestrian signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000022", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Debris scattered\", \"B\": \"Dry\", \"C\": \"Newly paved\", \"D\": \"Cracked asphalt\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000023", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000029.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000029.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Debris scattered\", \"B\": \"Dry\", \"C\": \"Newly paved\", \"D\": \"Cracked asphalt\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000023", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Subway entrance\", \"B\": \"Shopping district\", \"C\": \"Suburban\", \"D\": \"Airport roadway\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000024", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Subway entrance\", \"B\": \"Shopping district\", \"C\": \"Suburban\", \"D\": \"Airport roadway\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000024", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dusk\", \"B\": \"Low light\", \"C\": \"Glare\", \"D\": \"Smoky\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000025", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dusk\", \"B\": \"Low light\", \"C\": \"Glare\", \"D\": \"Smoky\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000025", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Heavy\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000026", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Heavy\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000026", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Drizzling\", \"B\": \"Sunny\", \"C\": \"Dusty\", \"D\": \"Overcast\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000027", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Drizzling\", \"B\": \"Sunny\", \"C\": \"Dusty\", \"D\": \"Overcast\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000027", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Temporary road sign light\", \"B\": \"None\", \"C\": \"Yellow\", \"D\": \"Reflected signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000028", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Temporary road sign light\", \"B\": \"None\", \"C\": \"Yellow\", \"D\": \"Reflected signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000028", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Snowy\", \"B\": \"Dry\", \"C\": \"Construction debris\", \"D\": \"Leaf-covered\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000029", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000037.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000037.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Snowy\", \"B\": \"Dry\", \"C\": \"Construction debris\", \"D\": \"Leaf-covered\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000029", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Forest road\", \"B\": \"Construction zone\", \"C\": \"Airport roadway\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000030", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Forest road\", \"B\": \"Construction zone\", \"C\": \"Airport roadway\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000030", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Low light\", \"C\": \"Under tunnel\", \"D\": \"Clear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000031", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Low light\", \"C\": \"Under tunnel\", \"D\": \"Clear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000031", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Heavy\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000032", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Heavy\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000032", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Snowstorm\", \"C\": \"Sunny\", \"D\": \"Mist\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000033", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Snowstorm\", \"C\": \"Sunny\", \"D\": \"Mist\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000033", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Temporary road sign light\", \"C\": \"Pedestrian signal\", \"D\": \"Flashing amber\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000034", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Temporary road sign light\", \"C\": \"Pedestrian signal\", \"D\": \"Flashing amber\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000034", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Bumpy\", \"B\": \"Dry\", \"C\": \"Oil slick\", \"D\": \"Sandy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000035", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000045.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000045.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Bumpy\", \"B\": \"Dry\", \"C\": \"Oil slick\", \"D\": \"Sandy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000035", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Service station\", \"C\": \"Construction zone\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000036", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Service station\", \"C\": \"Construction zone\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000036", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Low light\", \"C\": \"Reflected glare\", \"D\": \"Partly cloudy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000037", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Low light\", \"C\": \"Reflected glare\", \"D\": \"Partly cloudy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000037", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Vehicle queue\", \"C\": \"Light\", \"D\": \"Detour traffic\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000038", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Vehicle queue\", \"C\": \"Light\", \"D\": \"Detour traffic\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000038", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Sleet\", \"C\": \"Sunny\", \"D\": \"Smoggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000039", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Sleet\", \"C\": \"Sunny\", \"D\": \"Smoggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000039", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Temporary road sign light\", \"C\": \"Countdown timer visible\", \"D\": \"Pedestrian signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000040", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Temporary road sign light\", \"C\": \"Countdown timer visible\", \"D\": \"Pedestrian signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000040", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Water puddles\", \"B\": \"Bumpy\", \"C\": \"Dry\", \"D\": \"Rough surface\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000041", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000053.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000053.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Water puddles\", \"B\": \"Bumpy\", \"C\": \"Dry\", \"D\": \"Rough surface\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000041", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Harbor area\", \"B\": \"Urban\", \"C\": \"Construction zone\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000042", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Harbor area\", \"B\": \"Urban\", \"C\": \"Construction zone\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000042", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dim streetlight visibility\", \"B\": \"Under tunnel\", \"C\": \"Low light\", \"D\": \"Backlit\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000043", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dim streetlight visibility\", \"B\": \"Under tunnel\", \"C\": \"Low light\", \"D\": \"Backlit\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000043", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Heavy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000044", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Heavy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000044", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Cloudy\", \"B\": \"Sunny\", \"C\": \"Strong wind\", \"D\": \"Sandstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000045", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Cloudy\", \"B\": \"Sunny\", \"C\": \"Strong wind\", \"D\": \"Sandstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000045", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Countdown timer visible\", \"B\": \"Off / malfunctioning\", \"C\": \"None\", \"D\": \"Yellow\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000046", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Countdown timer visible\", \"B\": \"Off / malfunctioning\", \"C\": \"None\", \"D\": \"Yellow\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000046", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Partially damaged\", \"B\": \"Dry\", \"C\": \"Debris scattered\", \"D\": \"Worn markings\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000047", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000061.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000061.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Partially damaged\", \"B\": \"Dry\", \"C\": \"Debris scattered\", \"D\": \"Worn markings\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000047", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Shopping district\", \"C\": \"Parking lot\", \"D\": \"Bridge\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000048", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Shopping district\", \"C\": \"Parking lot\", \"D\": \"Bridge\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000048", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Blinding sunlight\", \"B\": \"Overexposed\", \"C\": \"Rain-blurred\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000049", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Blinding sunlight\", \"B\": \"Overexposed\", \"C\": \"Rain-blurred\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000049", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Light\", \"C\": \"Accident blockage\", \"D\": \"Vehicle queue\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000050", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Light\", \"C\": \"Accident blockage\", \"D\": \"Vehicle queue\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000050", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Dusty\", \"C\": \"Foggy\", \"D\": \"Gloomy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000051", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Dusty\", \"C\": \"Foggy\", \"D\": \"Gloomy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000051", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is a traffic light visible in the scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Off / malfunctioning\", \"C\": \"None\", \"D\": \"Pedestrian signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000052", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is a traffic light visible in the scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Off / malfunctioning\", \"C\": \"None\", \"D\": \"Pedestrian signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000052", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Rough surface\", \"B\": \"Dry\", \"C\": \"Icy\", \"D\": \"Gravel\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000053", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000069.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000069.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Rough surface\", \"B\": \"Dry\", \"C\": \"Icy\", \"D\": \"Gravel\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000053", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Shopping district\", \"B\": \"Urban\", \"C\": \"Construction zone\", \"D\": \"Highway\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000054", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Shopping district\", \"B\": \"Urban\", \"C\": \"Construction zone\", \"D\": \"Highway\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000054", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Under tunnel\", \"C\": \"Low light\", \"D\": \"Glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000055", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Under tunnel\", \"C\": \"Low light\", \"D\": \"Glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000055", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Rush hour\", \"B\": \"Light\", \"C\": \"Detour traffic\", \"D\": \"Moderate\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000056", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Rush hour\", \"B\": \"Light\", \"C\": \"Detour traffic\", \"D\": \"Moderate\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000056", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Strong wind\", \"B\": \"Thunderstorm\", \"C\": \"Sunny\", \"D\": \"Mist\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000057", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Strong wind\", \"B\": \"Thunderstorm\", \"C\": \"Sunny\", \"D\": \"Mist\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000057", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Dim / unclear\", \"C\": \"Pedestrian signal\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000058", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Dim / unclear\", \"C\": \"Pedestrian signal\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000058", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Blocked by obstacle\", \"B\": \"Dry\", \"C\": \"Gravel\", \"D\": \"Muddy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000059", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000077.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000077.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Blocked by obstacle\", \"B\": \"Dry\", \"C\": \"Gravel\", \"D\": \"Muddy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000059", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Parking lot\", \"B\": \"Highway\", \"C\": \"Urban\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000060", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Parking lot\", \"B\": \"Highway\", \"C\": \"Urban\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000060", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Partly cloudy\", \"C\": \"Low light\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000061", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Partly cloudy\", \"C\": \"Low light\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000061", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Detour traffic\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000062", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Detour traffic\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000062", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Smoggy\", \"C\": \"Snowy\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000063", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Smoggy\", \"C\": \"Snowy\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000063", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Flashing yellow\", \"B\": \"None\", \"C\": \"Red\", \"D\": \"Covered by tree\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000064", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Flashing yellow\", \"B\": \"None\", \"C\": \"Red\", \"D\": \"Covered by tree\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000064", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Worn markings\", \"B\": \"Snow-covered\", \"C\": \"Construction debris\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000065", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000085.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000085.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Worn markings\", \"B\": \"Snow-covered\", \"C\": \"Construction debris\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000065", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Campus area\", \"B\": \"Shopping district\", \"C\": \"Forest road\", \"D\": \"Urban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000066", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Campus area\", \"B\": \"Shopping district\", \"C\": \"Forest road\", \"D\": \"Urban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000066", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Rain-blurred\", \"C\": \"Flashing lights\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000067", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Rain-blurred\", \"C\": \"Flashing lights\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000067", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Detour traffic\", \"D\": \"Heavy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000068", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Detour traffic\", \"D\": \"Heavy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000068", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Snowstorm\", \"B\": \"Stormy\", \"C\": \"Thunderstorm\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000069", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Snowstorm\", \"B\": \"Stormy\", \"C\": \"Thunderstorm\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000069", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Yellow\", \"C\": \"None\", \"D\": \"Flashing red\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000070", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Yellow\", \"C\": \"None\", \"D\": \"Flashing red\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000070", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Rough surface\", \"B\": \"Cracked asphalt\", \"C\": \"Leaf-covered\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000071", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000093.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000093.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Rough surface\", \"B\": \"Cracked asphalt\", \"C\": \"Leaf-covered\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000071", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Forest road\", \"C\": \"Coastal road\", \"D\": \"Urban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000072", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Forest road\", \"C\": \"Coastal road\", \"D\": \"Urban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000072", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Backlit\", \"C\": \"Low light\", \"D\": \"Dawn\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000073", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Backlit\", \"C\": \"Low light\", \"D\": \"Dawn\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000073", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Lane merging\", \"B\": \"Rush hour\", \"C\": \"Moderate\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000074", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Lane merging\", \"B\": \"Rush hour\", \"C\": \"Moderate\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000074", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Flurries\", \"B\": \"Gloomy\", \"C\": \"Sunny\", \"D\": \"Stormy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000075", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Flurries\", \"B\": \"Gloomy\", \"C\": \"Sunny\", \"D\": \"Stormy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000075", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Flashing yellow\", \"C\": \"Flashing amber\", \"D\": \"Blinking green\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000076", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Flashing yellow\", \"C\": \"Flashing amber\", \"D\": \"Blinking green\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000076", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Water puddles\", \"B\": \"Wet\", \"C\": \"Snowy\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000077", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000101.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000101.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Water puddles\", \"B\": \"Wet\", \"C\": \"Snowy\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000077", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Market street\", \"C\": \"Rural\", \"D\": \"Construction zone\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000078", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Market street\", \"C\": \"Rural\", \"D\": \"Construction zone\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000078", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Flashing lights\", \"C\": \"Low light\", \"D\": \"Reflected glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000079", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Flashing lights\", \"C\": \"Low light\", \"D\": \"Reflected glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000079", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Light\", \"C\": \"Vehicle queue\", \"D\": \"Congested\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000080", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Light\", \"C\": \"Vehicle queue\", \"D\": \"Congested\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000080", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Smoggy\", \"B\": \"Sunny\", \"C\": \"Sandstorm\", \"D\": \"Cloudy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000081", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Smoggy\", \"B\": \"Sunny\", \"C\": \"Sandstorm\", \"D\": \"Cloudy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000081", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is a traffic light visible in the frame? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Blinking green\", \"C\": \"Flashing amber\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000082", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is a traffic light visible in the frame? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Blinking green\", \"C\": \"Flashing amber\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000082", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Wet\", \"C\": \"Flooded\", \"D\": \"Icy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000083", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000109.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000109.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Wet\", \"C\": \"Flooded\", \"D\": \"Icy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000083", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Rural\", \"B\": \"Suburban\", \"C\": \"Harbor area\", \"D\": \"Service station\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000084", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Rural\", \"B\": \"Suburban\", \"C\": \"Harbor area\", \"D\": \"Service station\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000084", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Under tunnel\", \"C\": \"Low light\", \"D\": \"Backlit\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000085", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Under tunnel\", \"C\": \"Low light\", \"D\": \"Backlit\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000085", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Lane merging\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Vehicle queue\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000086", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Lane merging\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Vehicle queue\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000086", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather condition is suggested by the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Snowstorm\", \"C\": \"Mist\", \"D\": \"Windy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000087", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather condition is suggested by the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Snowstorm\", \"C\": \"Mist\", \"D\": \"Windy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000087", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light visible on the right side of the road? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Countdown timer visible\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000088", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light visible on the right side of the road? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Countdown timer visible\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000088", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Slippery\", \"C\": \"Potholes\", \"D\": \"Cracked asphalt\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000089", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000117.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000117.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Slippery\", \"C\": \"Potholes\", \"D\": \"Cracked asphalt\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000089", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Parking lot\", \"C\": \"Subway entrance\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000090", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Parking lot\", \"C\": \"Subway entrance\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000090", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Smoky\", \"C\": \"Glare\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000091", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Smoky\", \"C\": \"Glare\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000091", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Vehicle queue\", \"C\": \"Congested\", \"D\": \"Accident blockage\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000092", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Vehicle queue\", \"C\": \"Congested\", \"D\": \"Accident blockage\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000092", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Stormy\", \"B\": \"Hailing\", \"C\": \"Sandstorm\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000093", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Stormy\", \"B\": \"Hailing\", \"C\": \"Sandstorm\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000093", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Flashing yellow\", \"C\": \"Reflected signal\", \"D\": \"Pedestrian signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000094", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Flashing yellow\", \"C\": \"Reflected signal\", \"D\": \"Pedestrian signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000094", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Sandy\", \"C\": \"Blocked by obstacle\", \"D\": \"Newly paved\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000095", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000125.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000125.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Sandy\", \"C\": \"Blocked by obstacle\", \"D\": \"Newly paved\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000095", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Suburban\", \"C\": \"Highway\", \"D\": \"Campus area\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000096", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Suburban\", \"C\": \"Highway\", \"D\": \"Campus area\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000096", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Smoky\", \"B\": \"Under tunnel\", \"C\": \"Strong shadows\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000097", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Smoky\", \"B\": \"Under tunnel\", \"C\": \"Strong shadows\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000097", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Vehicle queue\", \"C\": \"Accident blockage\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000098", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Vehicle queue\", \"C\": \"Accident blockage\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000098", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Drizzling\", \"C\": \"Gloomy\", \"D\": \"Mist\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000099", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Drizzling\", \"C\": \"Gloomy\", \"D\": \"Mist\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000099", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"None\", \"C\": \"Red\", \"D\": \"Blinking green\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000100", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"None\", \"C\": \"Red\", \"D\": \"Blinking green\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000100", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Wet\", \"B\": \"Dry\", \"C\": \"Icy\", \"D\": \"Gravel\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000101", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000133.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000133.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Wet\", \"B\": \"Dry\", \"C\": \"Icy\", \"D\": \"Gravel\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000101", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Shopping district\", \"B\": \"Suburban\", \"C\": \"Highway\", \"D\": \"Service station\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000102", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Shopping district\", \"B\": \"Suburban\", \"C\": \"Highway\", \"D\": \"Service station\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000102", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Under tunnel\", \"B\": \"Reflected glare\", \"C\": \"Low light\", \"D\": \"Hazy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000103", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Under tunnel\", \"B\": \"Reflected glare\", \"C\": \"Low light\", \"D\": \"Hazy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000103", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Accident blockage\", \"C\": \"Detour traffic\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000104", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Accident blockage\", \"C\": \"Detour traffic\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000104", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Hailing\", \"B\": \"Sunny\", \"C\": \"Snowstorm\", \"D\": \"Sandstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000105", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Hailing\", \"B\": \"Sunny\", \"C\": \"Snowstorm\", \"D\": \"Sandstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000105", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"None\", \"C\": \"Flashing amber\", \"D\": \"Blinking green\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000106", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"None\", \"C\": \"Flashing amber\", \"D\": \"Blinking green\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000106", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Oil slick\", \"B\": \"Leaf-covered\", \"C\": \"Slippery\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000107", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000141.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000141.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Oil slick\", \"B\": \"Leaf-covered\", \"C\": \"Slippery\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000107", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Campus area\", \"B\": \"Bridge\", \"C\": \"Suburban\", \"D\": \"Airport roadway\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000108", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Campus area\", \"B\": \"Bridge\", \"C\": \"Suburban\", \"D\": \"Airport roadway\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000108", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Under tunnel\", \"B\": \"Low light\", \"C\": \"Hazy\", \"D\": \"Overexposed\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000109", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Under tunnel\", \"B\": \"Low light\", \"C\": \"Hazy\", \"D\": \"Overexposed\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000109", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density on the road? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Lane merging\", \"C\": \"Moderate\", \"D\": \"Detour traffic\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000110", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density on the road? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Lane merging\", \"C\": \"Moderate\", \"D\": \"Detour traffic\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Moderate" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000110", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Gloomy\", \"B\": \"Sunny\", \"C\": \"Dusty\", \"D\": \"Mist\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000111", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Gloomy\", \"B\": \"Sunny\", \"C\": \"Dusty\", \"D\": \"Mist\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000111", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Dim / unclear\", \"C\": \"Countdown timer visible\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000112", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Dim / unclear\", \"C\": \"Countdown timer visible\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000112", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Icy\", \"C\": \"Gravel\", \"D\": \"Partially damaged\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000113", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000149.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000149.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Icy\", \"C\": \"Gravel\", \"D\": \"Partially damaged\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000113", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Suburban\", \"C\": \"Rural\", \"D\": \"Parking lot\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000114", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Suburban\", \"C\": \"Rural\", \"D\": \"Parking lot\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000114", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Flashing lights\", \"B\": \"Smoky\", \"C\": \"Hazy\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000115", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Flashing lights\", \"B\": \"Smoky\", \"C\": \"Hazy\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000115", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Heavy\", \"B\": \"Light\", \"C\": \"Lane merging\", \"D\": \"Vehicle queue\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000116", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Heavy\", \"B\": \"Light\", \"C\": \"Lane merging\", \"D\": \"Vehicle queue\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000116", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Smoggy\", \"C\": \"Hailing\", \"D\": \"Overcast\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000117", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Smoggy\", \"C\": \"Hailing\", \"D\": \"Overcast\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000117", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"None\", \"C\": \"Flashing red\", \"D\": \"Covered by tree\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000118", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"None\", \"C\": \"Flashing red\", \"D\": \"Covered by tree\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000118", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Bumpy\", \"B\": \"Construction debris\", \"C\": \"Rough surface\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000119", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000157.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000157.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Bumpy\", \"B\": \"Construction debris\", \"C\": \"Rough surface\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000119", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Forest road\", \"B\": \"Coastal road\", \"C\": \"Suburban\", \"D\": \"Urban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000120", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Forest road\", \"B\": \"Coastal road\", \"C\": \"Suburban\", \"D\": \"Urban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000120", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Flashing lights\", \"B\": \"Low light\", \"C\": \"Hazy\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000121", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Flashing lights\", \"B\": \"Low light\", \"C\": \"Hazy\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000121", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Lane merging\", \"C\": \"Light\", \"D\": \"Accident blockage\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000122", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Detour traffic\", \"B\": \"Lane merging\", \"C\": \"Light\", \"D\": \"Accident blockage\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000122", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Cloudy\", \"B\": \"Rainy\", \"C\": \"Mist\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000123", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Cloudy\", \"B\": \"Rainy\", \"C\": \"Mist\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000123", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Green\", \"C\": \"None\", \"D\": \"Countdown timer visible\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000124", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Green\", \"C\": \"None\", \"D\": \"Countdown timer visible\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000124", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Potholes\", \"B\": \"Cracked asphalt\", \"C\": \"Partially damaged\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000125", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000165.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000165.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Potholes\", \"B\": \"Cracked asphalt\", \"C\": \"Partially damaged\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000125", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Highway\", \"C\": \"Harbor area\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000126", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Highway\", \"C\": \"Harbor area\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000126", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Backlit\", \"C\": \"Overexposed\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000127", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Backlit\", \"C\": \"Overexposed\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000127", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Accident blockage\", \"D\": \"Heavy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000128", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Accident blockage\", \"D\": \"Heavy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000128", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are evident in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Strong wind\", \"C\": \"Rainy\", \"D\": \"Snowy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000129", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are evident in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Strong wind\", \"C\": \"Rainy\", \"D\": \"Snowy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000129", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic signal in this scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Covered by tree\", \"C\": \"Flashing amber\", \"D\": \"Pedestrian signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000130", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic signal in this scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Covered by tree\", \"C\": \"Flashing amber\", \"D\": \"Pedestrian signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000130", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Flooded\", \"B\": \"Rough surface\", \"C\": \"Dry\", \"D\": \"Slippery\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000131", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000173.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000173.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Flooded\", \"B\": \"Rough surface\", \"C\": \"Dry\", \"D\": \"Slippery\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000131", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Rural\", \"B\": \"Subway entrance\", \"C\": \"Suburban\", \"D\": \"Market street\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000132", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Rural\", \"B\": \"Subway entrance\", \"C\": \"Suburban\", \"D\": \"Market street\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000132", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Strong shadows\", \"C\": \"Low light\", \"D\": \"Flashing lights\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000133", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Strong shadows\", \"C\": \"Low light\", \"D\": \"Flashing lights\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000133", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Rush hour\", \"C\": \"Moderate\", \"D\": \"Congested\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000134", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Rush hour\", \"C\": \"Moderate\", \"D\": \"Congested\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000134", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Stormy\", \"C\": \"Smoggy\", \"D\": \"Snowy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000135", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Stormy\", \"C\": \"Smoggy\", \"D\": \"Snowy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000135", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Dim / unclear\", \"D\": \"Reflected signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000136", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Dim / unclear\", \"D\": \"Reflected signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000136", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Worn markings\", \"B\": \"Oil slick\", \"C\": \"Dry\", \"D\": \"Bumpy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000137", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000181.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000181.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Worn markings\", \"B\": \"Oil slick\", \"C\": \"Dry\", \"D\": \"Bumpy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000137", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Forest road\", \"C\": \"Highway\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000138", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Forest road\", \"C\": \"Highway\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000138", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Reflected glare\", \"C\": \"Clear\", \"D\": \"Foggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000139", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Reflected glare\", \"C\": \"Clear\", \"D\": \"Foggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000139", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Accident blockage\", \"C\": \"Heavy\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000140", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Accident blockage\", \"C\": \"Heavy\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000140", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Strong wind\", \"B\": \"Windy\", \"C\": \"Sunny\", \"D\": \"Smoggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000141", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Strong wind\", \"B\": \"Windy\", \"C\": \"Sunny\", \"D\": \"Smoggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000141", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Flashing red\", \"C\": \"Pedestrian signal\", \"D\": \"Flashing yellow\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000142", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Flashing red\", \"C\": \"Pedestrian signal\", \"D\": \"Flashing yellow\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000142", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Partially damaged\", \"B\": \"Dry\", \"C\": \"Blocked by obstacle\", \"D\": \"Gravel\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000143", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000189.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000189.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Partially damaged\", \"B\": \"Dry\", \"C\": \"Blocked by obstacle\", \"D\": \"Gravel\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000143", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Parking lot\", \"B\": \"Forest road\", \"C\": \"Suburban\", \"D\": \"Shopping district\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000144", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Parking lot\", \"B\": \"Forest road\", \"C\": \"Suburban\", \"D\": \"Shopping district\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000144", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Low light\", \"C\": \"Under tunnel\", \"D\": \"Smoky\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000145", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Low light\", \"C\": \"Under tunnel\", \"D\": \"Smoky\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000145", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Lane merging\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000146", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Lane merging\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000146", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Stormy\", \"C\": \"Mist\", \"D\": \"Foggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000147", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Stormy\", \"C\": \"Mist\", \"D\": \"Foggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000147", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the traffic light status from this perspective? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Flashing amber\", \"C\": \"Green\", \"D\": \"Off / malfunctioning\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000148", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the traffic light status from this perspective? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Flashing amber\", \"C\": \"Green\", \"D\": \"Off / malfunctioning\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000148", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Snowy\", \"B\": \"Dry\", \"C\": \"Leaf-covered\", \"D\": \"Water puddles\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000149", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000197.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000197.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Snowy\", \"B\": \"Dry\", \"C\": \"Leaf-covered\", \"D\": \"Water puddles\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000149", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Harbor area\", \"C\": \"Campus area\", \"D\": \"Shopping district\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000150", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Harbor area\", \"C\": \"Campus area\", \"D\": \"Shopping district\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000150", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Glare\", \"B\": \"Low light\", \"C\": \"Flashing lights\", \"D\": \"Strong shadows\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000151", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Glare\", \"B\": \"Low light\", \"C\": \"Flashing lights\", \"D\": \"Strong shadows\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000151", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Moderate\", \"C\": \"Accident blockage\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000152", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Moderate\", \"C\": \"Accident blockage\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000152", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Dusty\", \"C\": \"Overcast\", \"D\": \"Hailing\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000153", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Dusty\", \"C\": \"Overcast\", \"D\": \"Hailing\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000153", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"None\", \"C\": \"Off / malfunctioning\", \"D\": \"Yellow\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000154", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"None\", \"C\": \"Off / malfunctioning\", \"D\": \"Yellow\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000154", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Construction debris\", \"B\": \"Dry\", \"C\": \"Snow-covered\", \"D\": \"Sandy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000155", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000205.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000205.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Construction debris\", \"B\": \"Dry\", \"C\": \"Snow-covered\", \"D\": \"Sandy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000155", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Service station\", \"C\": \"Forest road\", \"D\": \"Campus area\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000156", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Service station\", \"C\": \"Forest road\", \"D\": \"Campus area\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000156", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Low light\", \"C\": \"Blinding sunlight\", \"D\": \"Under tunnel\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000157", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Low light\", \"C\": \"Blinding sunlight\", \"D\": \"Under tunnel\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000157", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Detour traffic\", \"D\": \"Moderate\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000158", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Detour traffic\", \"D\": \"Moderate\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000158", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Strong wind\", \"C\": \"Sleet\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000159", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Strong wind\", \"C\": \"Sleet\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000159", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in this scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Temporary road sign light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000160", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in this scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Temporary road sign light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000160", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Sandy\", \"C\": \"Wet\", \"D\": \"Cracked asphalt\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000161", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000221.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000221.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Sandy\", \"C\": \"Wet\", \"D\": \"Cracked asphalt\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000161", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Bridge\", \"C\": \"Forest road\", \"D\": \"Construction zone\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000162", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Bridge\", \"C\": \"Forest road\", \"D\": \"Construction zone\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000162", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Flashing lights\", \"B\": \"Glare\", \"C\": \"Rain-blurred\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000163", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Flashing lights\", \"B\": \"Glare\", \"C\": \"Rain-blurred\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000163", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Rush hour\", \"D\": \"Lane merging\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000164", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Rush hour\", \"D\": \"Lane merging\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000164", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Snowstorm\", \"B\": \"Sunny\", \"C\": \"Sleet\", \"D\": \"Drizzling\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000165", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Snowstorm\", \"B\": \"Sunny\", \"C\": \"Sleet\", \"D\": \"Drizzling\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000165", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic lights in this scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Flashing amber\", \"D\": \"Green\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000166", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic lights in this scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Flashing amber\", \"D\": \"Green\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000166", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Leaf-covered\", \"B\": \"Construction debris\", \"C\": \"Dry\", \"D\": \"Blocked by obstacle\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000167", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000229.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000229.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Leaf-covered\", \"B\": \"Construction debris\", \"C\": \"Dry\", \"D\": \"Blocked by obstacle\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000167", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Construction zone\", \"C\": \"Harbor area\", \"D\": \"Service station\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000168", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Construction zone\", \"C\": \"Harbor area\", \"D\": \"Service station\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000168", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Overexposed\", \"B\": \"Flashing lights\", \"C\": \"Smoky\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000169", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Overexposed\", \"B\": \"Flashing lights\", \"C\": \"Smoky\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000169", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Congested\", \"C\": \"Accident blockage\", \"D\": \"Moderate\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000170", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Congested\", \"C\": \"Accident blockage\", \"D\": \"Moderate\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000170", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Foggy\", \"C\": \"Snowy\", \"D\": \"Overcast\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000171", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Foggy\", \"C\": \"Snowy\", \"D\": \"Overcast\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000171", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Flashing amber\", \"C\": \"None\", \"D\": \"Dim / unclear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000172", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Flashing amber\", \"C\": \"None\", \"D\": \"Dim / unclear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000172", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Uneven\", \"C\": \"Oil slick\", \"D\": \"Slippery\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000173", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000237.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000237.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Uneven\", \"C\": \"Oil slick\", \"D\": \"Slippery\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000173", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Construction zone\", \"B\": \"Suburban\", \"C\": \"Urban\", \"D\": \"Rural\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000174", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Construction zone\", \"B\": \"Suburban\", \"C\": \"Urban\", \"D\": \"Rural\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000174", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Foggy\", \"C\": \"Smoky\", \"D\": \"Reflected glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000175", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Foggy\", \"C\": \"Smoky\", \"D\": \"Reflected glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000175", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Heavy\", \"C\": \"Accident blockage\", \"D\": \"Detour traffic\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000176", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Heavy\", \"C\": \"Accident blockage\", \"D\": \"Detour traffic\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000176", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Foggy\", \"C\": \"Windy\", \"D\": \"Dusty\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000177", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Foggy\", \"C\": \"Windy\", \"D\": \"Dusty\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000177", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"None\", \"C\": \"Countdown timer visible\", \"D\": \"Temporary road sign light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000178", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"None\", \"C\": \"Countdown timer visible\", \"D\": \"Temporary road sign light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000178", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Partially damaged\", \"B\": \"Snow-covered\", \"C\": \"Dry\", \"D\": \"Snowy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000179", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000245.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000245.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Partially damaged\", \"B\": \"Snow-covered\", \"C\": \"Dry\", \"D\": \"Snowy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000179", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Campus area\", \"B\": \"Coastal road\", \"C\": \"Suburban\", \"D\": \"Subway entrance\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000180", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Campus area\", \"B\": \"Coastal road\", \"C\": \"Suburban\", \"D\": \"Subway entrance\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000180", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Low light\", \"C\": \"Dim streetlight visibility\", \"D\": \"Overexposed\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000181", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Low light\", \"C\": \"Dim streetlight visibility\", \"D\": \"Overexposed\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000181", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Accident blockage\", \"C\": \"Lane merging\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000182", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Accident blockage\", \"C\": \"Lane merging\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000182", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Rainy\", \"B\": \"Sunny\", \"C\": \"Partly cloudy\", \"D\": \"Cloudy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000183", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Rainy\", \"B\": \"Sunny\", \"C\": \"Partly cloudy\", \"D\": \"Cloudy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000183", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Temporary road sign light\", \"D\": \"Countdown timer visible\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000184", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Temporary road sign light\", \"D\": \"Countdown timer visible\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000184", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Worn markings\", \"B\": \"Dry\", \"C\": \"Flooded\", \"D\": \"Wet\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000185", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000253.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000253.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Worn markings\", \"B\": \"Dry\", \"C\": \"Flooded\", \"D\": \"Wet\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000185", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Suburban\", \"C\": \"Campus area\", \"D\": \"Parking lot\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000186", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Suburban\", \"C\": \"Campus area\", \"D\": \"Parking lot\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000186", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Smoky\", \"B\": \"Low light\", \"C\": \"Strong shadows\", \"D\": \"Reflected glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000187", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Smoky\", \"B\": \"Low light\", \"C\": \"Strong shadows\", \"D\": \"Reflected glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000187", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Vehicle queue\", \"C\": \"Light\", \"D\": \"Moderate\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000188", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Vehicle queue\", \"C\": \"Light\", \"D\": \"Moderate\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000188", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Sleet\", \"C\": \"Drizzling\", \"D\": \"Thunderstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000189", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Sleet\", \"C\": \"Drizzling\", \"D\": \"Thunderstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000189", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the frame? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Countdown timer visible\", \"C\": \"Reflected signal\", \"D\": \"Flashing red\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000190", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the frame? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Countdown timer visible\", \"C\": \"Reflected signal\", \"D\": \"Flashing red\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000190", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Debris scattered\", \"B\": \"Dry\", \"C\": \"Oil slick\", \"D\": \"Bumpy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000191", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000469.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000469.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Debris scattered\", \"B\": \"Dry\", \"C\": \"Oil slick\", \"D\": \"Bumpy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000191", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Rural\", \"B\": \"Campus area\", \"C\": \"Suburban\", \"D\": \"Parking lot\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000192", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Rural\", \"B\": \"Campus area\", \"C\": \"Suburban\", \"D\": \"Parking lot\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000192", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Under tunnel\", \"B\": \"Glare\", \"C\": \"Low light\", \"D\": \"Strong shadows\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000193", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Under tunnel\", \"B\": \"Glare\", \"C\": \"Low light\", \"D\": \"Strong shadows\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000193", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Lane merging\", \"C\": \"Detour traffic\", \"D\": \"Heavy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000194", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Lane merging\", \"C\": \"Detour traffic\", \"D\": \"Heavy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000194", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather condition is suggested by the scene? You should answer one choice from four answers \"A\": \"Smoggy\", \"B\": \"Flurries\", \"C\": \"Overcast\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000195", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather condition is suggested by the scene? You should answer one choice from four answers \"A\": \"Smoggy\", \"B\": \"Flurries\", \"C\": \"Overcast\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000195", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Green\", \"C\": \"Flashing red\", \"D\": \"Off / malfunctioning\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000196", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Green\", \"C\": \"Flashing red\", \"D\": \"Off / malfunctioning\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000196", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Rough surface\", \"B\": \"Dry\", \"C\": \"Water puddles\", \"D\": \"Snowy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000197", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000477.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000477.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Rough surface\", \"B\": \"Dry\", \"C\": \"Water puddles\", \"D\": \"Snowy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000197", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Forest road\", \"B\": \"Airport roadway\", \"C\": \"Urban\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000198", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Forest road\", \"B\": \"Airport roadway\", \"C\": \"Urban\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000198", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dawn\", \"B\": \"Clear\", \"C\": \"Low light\", \"D\": \"Overexposed\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000199", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dawn\", \"B\": \"Clear\", \"C\": \"Low light\", \"D\": \"Overexposed\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000199", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Lane merging\", \"D\": \"Heavy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000200", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Lane merging\", \"D\": \"Heavy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000200", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Overcast\", \"B\": \"Thunderstorm\", \"C\": \"Snowstorm\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000201", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Overcast\", \"B\": \"Thunderstorm\", \"C\": \"Snowstorm\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000201", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light visible in the distance? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Countdown timer visible\", \"C\": \"Red\", \"D\": \"Blinking green\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000202", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light visible in the distance? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Countdown timer visible\", \"C\": \"Red\", \"D\": \"Blinking green\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000202", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Bumpy\", \"B\": \"Gravel\", \"C\": \"Newly paved\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000203", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000485.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000485.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Bumpy\", \"B\": \"Gravel\", \"C\": \"Newly paved\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000203", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Highway\", \"B\": \"Suburban\", \"C\": \"Bridge\", \"D\": \"Coastal road\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000204", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Highway\", \"B\": \"Suburban\", \"C\": \"Bridge\", \"D\": \"Coastal road\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000204", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Low light\", \"C\": \"Flashing lights\", \"D\": \"Glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000205", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Low light\", \"C\": \"Flashing lights\", \"D\": \"Glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000205", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Rush hour\", \"B\": \"Vehicle queue\", \"C\": \"Accident blockage\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000206", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Rush hour\", \"B\": \"Vehicle queue\", \"C\": \"Accident blockage\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000206", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Snowy\", \"B\": \"Sunny\", \"C\": \"Hailing\", \"D\": \"Smoggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000207", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Snowy\", \"B\": \"Sunny\", \"C\": \"Hailing\", \"D\": \"Smoggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000207", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light visible in the distance? You should answer one choice from four answers \"A\": \"Dim / unclear\", \"B\": \"Temporary road sign light\", \"C\": \"Countdown timer visible\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000208", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light visible in the distance? You should answer one choice from four answers \"A\": \"Dim / unclear\", \"B\": \"Temporary road sign light\", \"C\": \"Countdown timer visible\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000208", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in this scene? You should answer one choice from four answers \"A\": \"Construction debris\", \"B\": \"Dry\", \"C\": \"Worn markings\", \"D\": \"Sandy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000209", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000493.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000493.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in this scene? You should answer one choice from four answers \"A\": \"Construction debris\", \"B\": \"Dry\", \"C\": \"Worn markings\", \"D\": \"Sandy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000209", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Subway entrance\", \"B\": \"Suburban\", \"C\": \"Forest road\", \"D\": \"Intersection\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000210", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Subway entrance\", \"B\": \"Suburban\", \"C\": \"Forest road\", \"D\": \"Intersection\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000210", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Hazy\", \"C\": \"Dawn\", \"D\": \"Clear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000211", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Hazy\", \"C\": \"Dawn\", \"D\": \"Clear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000211", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Congested\", \"D\": \"Accident blockage\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000212", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Congested\", \"D\": \"Accident blockage\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000212", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Stormy\", \"B\": \"Cloudy\", \"C\": \"Sunny\", \"D\": \"Mist\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000213", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Stormy\", \"B\": \"Cloudy\", \"C\": \"Sunny\", \"D\": \"Mist\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000213", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic light in the scene? You should answer one choice from four answers \"A\": \"Pedestrian signal\", \"B\": \"None\", \"C\": \"Green\", \"D\": \"Flashing red\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000214", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic light in the scene? You should answer one choice from four answers \"A\": \"Pedestrian signal\", \"B\": \"None\", \"C\": \"Green\", \"D\": \"Flashing red\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000214", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Uneven\", \"B\": \"Snow-covered\", \"C\": \"Bumpy\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000215", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000501.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000501.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Uneven\", \"B\": \"Snow-covered\", \"C\": \"Bumpy\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000215", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Harbor area\", \"C\": \"Urban\", \"D\": \"Construction zone\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000216", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Harbor area\", \"C\": \"Urban\", \"D\": \"Construction zone\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000216", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Under tunnel\", \"B\": \"Low light\", \"C\": \"Strong shadows\", \"D\": \"Glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000217", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Under tunnel\", \"B\": \"Low light\", \"C\": \"Strong shadows\", \"D\": \"Glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000217", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Rush hour\", \"B\": \"Accident blockage\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000218", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Rush hour\", \"B\": \"Accident blockage\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000218", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Mist\", \"B\": \"Strong wind\", \"C\": \"Sunny\", \"D\": \"Sleet\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000219", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Mist\", \"B\": \"Strong wind\", \"C\": \"Sunny\", \"D\": \"Sleet\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000219", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light visible in the distance? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Dim / unclear\", \"C\": \"Pedestrian signal\", \"D\": \"Flashing red\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000220", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light visible in the distance? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Dim / unclear\", \"C\": \"Pedestrian signal\", \"D\": \"Flashing red\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000220", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Potholes\", \"C\": \"Wet\", \"D\": \"Worn markings\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000221", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000509.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000509.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Potholes\", \"C\": \"Wet\", \"D\": \"Worn markings\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000221", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Forest road\", \"C\": \"Suburban\", \"D\": \"Market street\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000222", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Forest road\", \"C\": \"Suburban\", \"D\": \"Market street\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000222", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Dawn\", \"C\": \"Low light\", \"D\": \"Blinding sunlight\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000223", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Dawn\", \"C\": \"Low light\", \"D\": \"Blinding sunlight\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000223", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Vehicle queue\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000224", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Vehicle queue\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000224", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Gloomy\", \"B\": \"Flurries\", \"C\": \"Sunny\", \"D\": \"Sleet\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000225", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Gloomy\", \"B\": \"Flurries\", \"C\": \"Sunny\", \"D\": \"Sleet\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000225", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in this frame? You should answer one choice from four answers \"A\": \"Blinking green\", \"B\": \"Covered by tree\", \"C\": \"Yellow\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000226", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in this frame? You should answer one choice from four answers \"A\": \"Blinking green\", \"B\": \"Covered by tree\", \"C\": \"Yellow\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000226", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Water puddles\", \"B\": \"Wet\", \"C\": \"Worn markings\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000227", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000541.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000541.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Water puddles\", \"B\": \"Wet\", \"C\": \"Worn markings\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000227", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Bridge\", \"C\": \"Parking lot\", \"D\": \"Construction zone\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000228", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Bridge\", \"C\": \"Parking lot\", \"D\": \"Construction zone\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000228", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Overexposed\", \"B\": \"Low light\", \"C\": \"Dim streetlight visibility\", \"D\": \"Reflected glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000229", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Overexposed\", \"B\": \"Low light\", \"C\": \"Dim streetlight visibility\", \"D\": \"Reflected glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000229", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density on the road? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Heavy\", \"C\": \"Rush hour\", \"D\": \"Lane merging\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000230", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density on the road? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Heavy\", \"C\": \"Rush hour\", \"D\": \"Lane merging\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000230", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the current weather conditions? You should answer one choice from four answers \"A\": \"Windy\", \"B\": \"Sunny\", \"C\": \"Dusty\", \"D\": \"Flurries\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000231", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the current weather conditions? You should answer one choice from four answers \"A\": \"Windy\", \"B\": \"Sunny\", \"C\": \"Dusty\", \"D\": \"Flurries\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000231", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light ahead? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"Countdown timer visible\", \"C\": \"Red\", \"D\": \"Dim / unclear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000232", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the state of the traffic light ahead? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"Countdown timer visible\", \"C\": \"Red\", \"D\": \"Dim / unclear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Red" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000232", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Construction debris\", \"B\": \"Dry\", \"C\": \"Snowy\", \"D\": \"Water puddles\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000233", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000549.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000549.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Construction debris\", \"B\": \"Dry\", \"C\": \"Snowy\", \"D\": \"Water puddles\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000233", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Roundabout\", \"B\": \"Suburban\", \"C\": \"Subway entrance\", \"D\": \"Service station\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000234", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Roundabout\", \"B\": \"Suburban\", \"C\": \"Subway entrance\", \"D\": \"Service station\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000234", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Foggy\", \"C\": \"Low light\", \"D\": \"Glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000235", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Foggy\", \"C\": \"Low light\", \"D\": \"Glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000235", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Light\", \"C\": \"Lane merging\", \"D\": \"Rush hour\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000236", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Light\", \"C\": \"Lane merging\", \"D\": \"Rush hour\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000236", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Overcast\", \"B\": \"Sandstorm\", \"C\": \"Smoggy\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000237", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Overcast\", \"B\": \"Sandstorm\", \"C\": \"Smoggy\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000237", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in this scene? You should answer one choice from four answers \"A\": \"Pedestrian signal\", \"B\": \"None\", \"C\": \"Temporary road sign light\", \"D\": \"Blinking green\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000238", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in this scene? You should answer one choice from four answers \"A\": \"Pedestrian signal\", \"B\": \"None\", \"C\": \"Temporary road sign light\", \"D\": \"Blinking green\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000238", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Bumpy\", \"C\": \"Icy\", \"D\": \"Snow-covered\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000239", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000557.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000557.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Bumpy\", \"C\": \"Icy\", \"D\": \"Snow-covered\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000239", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Coastal road\", \"C\": \"Airport roadway\", \"D\": \"Intersection\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000240", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Coastal road\", \"C\": \"Airport roadway\", \"D\": \"Intersection\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000240", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Dim streetlight visibility\", \"B\": \"Low light\", \"C\": \"Dawn\", \"D\": \"Flashing lights\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000241", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Dim streetlight visibility\", \"B\": \"Low light\", \"C\": \"Dawn\", \"D\": \"Flashing lights\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000241", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Vehicle queue\", \"C\": \"Detour traffic\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000242", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Vehicle queue\", \"C\": \"Detour traffic\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000242", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Flurries\", \"B\": \"Rainy\", \"C\": \"Sunny\", \"D\": \"Windy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000243", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Flurries\", \"B\": \"Rainy\", \"C\": \"Sunny\", \"D\": \"Windy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000243", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the visible signal state of the traffic light in the distance? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Covered by tree\", \"C\": \"Temporary road sign light\", \"D\": \"Dim / unclear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000244", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the visible signal state of the traffic light in the distance? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Covered by tree\", \"C\": \"Temporary road sign light\", \"D\": \"Dim / unclear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Red" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000244", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Leaf-covered\", \"C\": \"Debris scattered\", \"D\": \"Water puddles\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000245", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000565.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000565.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Leaf-covered\", \"C\": \"Debris scattered\", \"D\": \"Water puddles\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000245", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Construction zone\", \"C\": \"Bridge\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000246", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Construction zone\", \"C\": \"Bridge\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000246", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Dusk\", \"C\": \"Low light\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000247", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Dusk\", \"C\": \"Low light\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000247", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Vehicle queue\", \"C\": \"Rush hour\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000248", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Vehicle queue\", \"C\": \"Rush hour\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000248", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in this scene? You should answer one choice from four answers \"A\": \"Flurries\", \"B\": \"Thunderstorm\", \"C\": \"Sunny\", \"D\": \"Gloomy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000249", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in this scene? You should answer one choice from four answers \"A\": \"Flurries\", \"B\": \"Thunderstorm\", \"C\": \"Sunny\", \"D\": \"Gloomy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000249", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is a traffic light visible in the frame? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Dim / unclear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000250", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is a traffic light visible in the frame? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Dim / unclear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000250", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Flooded\", \"C\": \"Cracked asphalt\", \"D\": \"Uneven\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000251", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000573.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000573.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Flooded\", \"C\": \"Cracked asphalt\", \"D\": \"Uneven\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000251", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Highway\", \"B\": \"Suburban\", \"C\": \"Construction zone\", \"D\": \"Subway entrance\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000252", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Highway\", \"B\": \"Suburban\", \"C\": \"Construction zone\", \"D\": \"Subway entrance\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000252", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Smoky\", \"C\": \"Glare\", \"D\": \"Clear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000253", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Smoky\", \"C\": \"Glare\", \"D\": \"Clear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000253", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Vehicle queue\", \"D\": \"Detour traffic\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000254", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Vehicle queue\", \"D\": \"Detour traffic\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000254", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are described in the scene? You should answer one choice from four answers \"A\": \"Partly cloudy\", \"B\": \"Sunny\", \"C\": \"Sleet\", \"D\": \"Snowstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000255", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are described in the scene? You should answer one choice from four answers \"A\": \"Partly cloudy\", \"B\": \"Sunny\", \"C\": \"Sleet\", \"D\": \"Snowstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000255", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Are there any traffic lights visible in the scene? You should answer one choice from four answers \"A\": \"Flashing amber\", \"B\": \"Yellow\", \"C\": \"Dim / unclear\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000256", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Are there any traffic lights visible in the scene? You should answer one choice from four answers \"A\": \"Flashing amber\", \"B\": \"Yellow\", \"C\": \"Dim / unclear\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000256", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Oil slick\", \"C\": \"Bumpy\", \"D\": \"Muddy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000257", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000581.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000581.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Oil slick\", \"C\": \"Bumpy\", \"D\": \"Muddy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000257", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Urban\", \"C\": \"Bridge\", \"D\": \"Parking lot\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000258", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Urban\", \"C\": \"Bridge\", \"D\": \"Parking lot\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000258", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Reflected glare\", \"C\": \"Foggy\", \"D\": \"Strong shadows\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000259", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Reflected glare\", \"C\": \"Foggy\", \"D\": \"Strong shadows\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000259", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Moderate\", \"C\": \"Congested\", \"D\": \"Accident blockage\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000260", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Moderate\", \"C\": \"Congested\", \"D\": \"Accident blockage\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000260", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Stormy\", \"C\": \"Flurries\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000261", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Foggy\", \"B\": \"Stormy\", \"C\": \"Flurries\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000261", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Flashing red\", \"B\": \"None\", \"C\": \"Blinking green\", \"D\": \"Dim / unclear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000262", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Flashing red\", \"B\": \"None\", \"C\": \"Blinking green\", \"D\": \"Dim / unclear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000262", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Slippery\", \"B\": \"Dry\", \"C\": \"Cracked asphalt\", \"D\": \"Wet\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000263", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000589.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000589.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Slippery\", \"B\": \"Dry\", \"C\": \"Cracked asphalt\", \"D\": \"Wet\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000263", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Subway entrance\", \"B\": \"Intersection\", \"C\": \"Urban\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000264", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Subway entrance\", \"B\": \"Intersection\", \"C\": \"Urban\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000264", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Low light\", \"C\": \"Backlit\", \"D\": \"Rain-blurred\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000265", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Clear\", \"B\": \"Low light\", \"C\": \"Backlit\", \"D\": \"Rain-blurred\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000265", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Detour traffic\", \"C\": \"Moderate\", \"D\": \"Vehicle queue\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000266", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Detour traffic\", \"C\": \"Moderate\", \"D\": \"Vehicle queue\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000266", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Drizzling\", \"B\": \"Gloomy\", \"C\": \"Snowy\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000267", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Drizzling\", \"B\": \"Gloomy\", \"C\": \"Snowy\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000267", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Pedestrian signal\", \"B\": \"None\", \"C\": \"Reflected signal\", \"D\": \"Temporary road sign light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000268", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Pedestrian signal\", \"B\": \"None\", \"C\": \"Reflected signal\", \"D\": \"Temporary road sign light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000268", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Newly paved\", \"B\": \"Worn markings\", \"C\": \"Gravel\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000269", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000597.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000597.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Newly paved\", \"B\": \"Worn markings\", \"C\": \"Gravel\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000269", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Roundabout\", \"C\": \"Rural\", \"D\": \"Parking lot\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000270", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Roundabout\", \"C\": \"Rural\", \"D\": \"Parking lot\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000270", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Dusk\", \"C\": \"Dim streetlight visibility\", \"D\": \"Clear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000271", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Dusk\", \"C\": \"Dim streetlight visibility\", \"D\": \"Clear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000271", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Accident blockage\", \"C\": \"Moderate\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000272", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Accident blockage\", \"C\": \"Moderate\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000272", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Thunderstorm\", \"B\": \"Sunny\", \"C\": \"Strong wind\", \"D\": \"Snowy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000273", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Thunderstorm\", \"B\": \"Sunny\", \"C\": \"Strong wind\", \"D\": \"Snowy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000273", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Pedestrian signal\", \"D\": \"Flashing red\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000274", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Pedestrian signal\", \"D\": \"Flashing red\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000274", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Blocked by obstacle\", \"C\": \"Oil slick\", \"D\": \"Newly paved\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000275", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000605.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000605.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Blocked by obstacle\", \"C\": \"Oil slick\", \"D\": \"Newly paved\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000275", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Coastal road\", \"C\": \"Construction zone\", \"D\": \"Parking lot\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000276", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Coastal road\", \"C\": \"Construction zone\", \"D\": \"Parking lot\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000276", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Strong shadows\", \"C\": \"Dawn\", \"D\": \"Clear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000277", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Strong shadows\", \"C\": \"Dawn\", \"D\": \"Clear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000277", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Rush hour\", \"C\": \"Light\", \"D\": \"Detour traffic\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000278", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Rush hour\", \"C\": \"Light\", \"D\": \"Detour traffic\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000278", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Thunderstorm\", \"C\": \"Stormy\", \"D\": \"Smoggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000279", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Thunderstorm\", \"C\": \"Stormy\", \"D\": \"Smoggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000279", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Covered by tree\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000280", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Covered by tree\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000280", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Potholes\", \"B\": \"Cracked asphalt\", \"C\": \"Dry\", \"D\": \"Wet\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000281", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000613.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000613.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Potholes\", \"B\": \"Cracked asphalt\", \"C\": \"Dry\", \"D\": \"Wet\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000281", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Bridge\", \"C\": \"Intersection\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000282", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Coastal road\", \"B\": \"Bridge\", \"C\": \"Intersection\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000282", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Clear\", \"C\": \"Glare\", \"D\": \"Reflected glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000283", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Clear\", \"C\": \"Glare\", \"D\": \"Reflected glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000283", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Lane merging\", \"B\": \"Rush hour\", \"C\": \"Light\", \"D\": \"Heavy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000284", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Lane merging\", \"B\": \"Rush hour\", \"C\": \"Light\", \"D\": \"Heavy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000284", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Smoggy\", \"B\": \"Strong wind\", \"C\": \"Sunny\", \"D\": \"Snowstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000285", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Smoggy\", \"B\": \"Strong wind\", \"C\": \"Sunny\", \"D\": \"Snowstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000285", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Temporary road sign light\", \"B\": \"Blinking green\", \"C\": \"Covered by tree\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000286", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Temporary road sign light\", \"B\": \"Blinking green\", \"C\": \"Covered by tree\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000286", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Cracked asphalt\", \"B\": \"Dry\", \"C\": \"Partially damaged\", \"D\": \"Sandy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000287", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000621.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000621.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Cracked asphalt\", \"B\": \"Dry\", \"C\": \"Partially damaged\", \"D\": \"Sandy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000287", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Construction zone\", \"B\": \"Forest road\", \"C\": \"Suburban\", \"D\": \"Urban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000288", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Construction zone\", \"B\": \"Forest road\", \"C\": \"Suburban\", \"D\": \"Urban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000288", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Partly cloudy\", \"B\": \"Rain-blurred\", \"C\": \"Low light\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000289", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Partly cloudy\", \"B\": \"Rain-blurred\", \"C\": \"Low light\", \"D\": \"Dim streetlight visibility\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000289", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Moderate\", \"C\": \"Vehicle queue\", \"D\": \"Detour traffic\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000290", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Moderate\", \"C\": \"Vehicle queue\", \"D\": \"Detour traffic\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000290", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Gloomy\", \"C\": \"Sleet\", \"D\": \"Snowstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000291", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Gloomy\", \"C\": \"Sleet\", \"D\": \"Snowstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000291", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Red\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000292", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"Flashing red\", \"C\": \"None\", \"D\": \"Red\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000292", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Potholes\", \"B\": \"Gravel\", \"C\": \"Icy\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000293", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000629.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000629.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Potholes\", \"B\": \"Gravel\", \"C\": \"Icy\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000293", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Roundabout\", \"C\": \"Urban\", \"D\": \"Bridge\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000294", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Roundabout\", \"C\": \"Urban\", \"D\": \"Bridge\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000294", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Hazy\", \"B\": \"Low light\", \"C\": \"Reflected glare\", \"D\": \"Dusk\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000295", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Hazy\", \"B\": \"Low light\", \"C\": \"Reflected glare\", \"D\": \"Dusk\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000295", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Lane merging\", \"C\": \"Moderate\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000296", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Congested\", \"B\": \"Lane merging\", \"C\": \"Moderate\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000296", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Rainy\", \"B\": \"Dusty\", \"C\": \"Mist\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000297", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Rainy\", \"B\": \"Dusty\", \"C\": \"Mist\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000297", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is a traffic light visible in the frame? You should answer one choice from four answers \"A\": \"Blinking green\", \"B\": \"Off / malfunctioning\", \"C\": \"None\", \"D\": \"Temporary road sign light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000298", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is a traffic light visible in the frame? You should answer one choice from four answers \"A\": \"Blinking green\", \"B\": \"Off / malfunctioning\", \"C\": \"None\", \"D\": \"Temporary road sign light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000298", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Rough surface\", \"C\": \"Snow-covered\", \"D\": \"Cracked asphalt\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000299", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000637.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000637.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Rough surface\", \"C\": \"Snow-covered\", \"D\": \"Cracked asphalt\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000299", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Intersection\", \"B\": \"Market street\", \"C\": \"Subway entrance\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000300", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Intersection\", \"B\": \"Market street\", \"C\": \"Subway entrance\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000300", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Low light\", \"C\": \"Under tunnel\", \"D\": \"Flashing lights\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000301", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Low light\", \"C\": \"Under tunnel\", \"D\": \"Flashing lights\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000301", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Heavy\", \"B\": \"Detour traffic\", \"C\": \"Light\", \"D\": \"Vehicle queue\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000302", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Heavy\", \"B\": \"Detour traffic\", \"C\": \"Light\", \"D\": \"Vehicle queue\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000302", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Thunderstorm\", \"C\": \"Dusty\", \"D\": \"Mist\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000303", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Thunderstorm\", \"C\": \"Dusty\", \"D\": \"Mist\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000303", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Flashing yellow\", \"B\": \"None\", \"C\": \"Red\", \"D\": \"Flashing red\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000304", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Flashing yellow\", \"B\": \"None\", \"C\": \"Red\", \"D\": \"Flashing red\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000304", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Snowy\", \"B\": \"Flooded\", \"C\": \"Oil slick\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000305", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000645.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000645.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Snowy\", \"B\": \"Flooded\", \"C\": \"Oil slick\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000305", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Shopping district\", \"C\": \"Urban\", \"D\": \"Rural\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000306", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Shopping district\", \"C\": \"Urban\", \"D\": \"Rural\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000306", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Dusk\", \"B\": \"Low light\", \"C\": \"Foggy\", \"D\": \"Dawn\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000307", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Dusk\", \"B\": \"Low light\", \"C\": \"Foggy\", \"D\": \"Dawn\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000307", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Light\", \"C\": \"Congested\", \"D\": \"Rush hour\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000308", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Light\", \"C\": \"Congested\", \"D\": \"Rush hour\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000308", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Cloudy\", \"B\": \"Sunny\", \"C\": \"Foggy\", \"D\": \"Snowstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000309", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Cloudy\", \"B\": \"Sunny\", \"C\": \"Foggy\", \"D\": \"Snowstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000309", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"Blinking green\", \"C\": \"Countdown timer visible\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000310", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Yellow\", \"B\": \"Blinking green\", \"C\": \"Countdown timer visible\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000310", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Worn markings\", \"B\": \"Oil slick\", \"C\": \"Slippery\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000311", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000653.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000653.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Worn markings\", \"B\": \"Oil slick\", \"C\": \"Slippery\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000311", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Harbor area\", \"B\": \"Suburban\", \"C\": \"Service station\", \"D\": \"Airport roadway\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000312", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Harbor area\", \"B\": \"Suburban\", \"C\": \"Service station\", \"D\": \"Airport roadway\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000312", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Smoky\", \"C\": \"Strong shadows\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000313", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Rain-blurred\", \"B\": \"Smoky\", \"C\": \"Strong shadows\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000313", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Lane merging\", \"C\": \"Congested\", \"D\": \"Detour traffic\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000314", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Lane merging\", \"C\": \"Congested\", \"D\": \"Detour traffic\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000314", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Windy\", \"C\": \"Strong wind\", \"D\": \"Gloomy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000315", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Windy\", \"C\": \"Strong wind\", \"D\": \"Gloomy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000315", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Green\", \"B\": \"Blinking green\", \"C\": \"None\", \"D\": \"Covered by tree\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000316", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Green\", \"B\": \"Blinking green\", \"C\": \"None\", \"D\": \"Covered by tree\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000316", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in the scene? You should answer one choice from four answers \"A\": \"Debris scattered\", \"B\": \"Dry\", \"C\": \"Flooded\", \"D\": \"Potholes\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000317", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000661.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000661.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in the scene? You should answer one choice from four answers \"A\": \"Debris scattered\", \"B\": \"Dry\", \"C\": \"Flooded\", \"D\": \"Potholes\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000317", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Subway entrance\", \"C\": \"Market street\", \"D\": \"Bridge\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000318", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Suburban\", \"B\": \"Subway entrance\", \"C\": \"Market street\", \"D\": \"Bridge\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000318", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Overexposed\", \"B\": \"Rain-blurred\", \"C\": \"Low light\", \"D\": \"Dawn\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000319", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Overexposed\", \"B\": \"Rain-blurred\", \"C\": \"Low light\", \"D\": \"Dawn\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000319", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Accident blockage\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000320", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Vehicle queue\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Accident blockage\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000320", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Snowstorm\", \"B\": \"Dusty\", \"C\": \"Sunny\", \"D\": \"Thunderstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000321", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Snowstorm\", \"B\": \"Dusty\", \"C\": \"Sunny\", \"D\": \"Thunderstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000321", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Are there any traffic lights visible in the scene? You should answer one choice from four answers \"A\": \"Flashing red\", \"B\": \"Temporary road sign light\", \"C\": \"Dim / unclear\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000322", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Are there any traffic lights visible in the scene? You should answer one choice from four answers \"A\": \"Flashing red\", \"B\": \"Temporary road sign light\", \"C\": \"Dim / unclear\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000322", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Oil slick\", \"B\": \"Muddy\", \"C\": \"Dry\", \"D\": \"Worn markings\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000323", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000669.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000669.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Oil slick\", \"B\": \"Muddy\", \"C\": \"Dry\", \"D\": \"Worn markings\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000323", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Forest road\", \"B\": \"Bridge\", \"C\": \"Suburban\", \"D\": \"Market street\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000324", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Forest road\", \"B\": \"Bridge\", \"C\": \"Suburban\", \"D\": \"Market street\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000324", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Overexposed\", \"B\": \"Backlit\", \"C\": \"Clear\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000325", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Overexposed\", \"B\": \"Backlit\", \"C\": \"Clear\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000325", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Moderate\", \"C\": \"Accident blockage\", \"D\": \"Rush hour\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000326", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Moderate\", \"C\": \"Accident blockage\", \"D\": \"Rush hour\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000326", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Hailing\", \"C\": \"Snowstorm\", \"D\": \"Drizzling\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000327", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Hailing\", \"C\": \"Snowstorm\", \"D\": \"Drizzling\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000327", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the image? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Blinking green\", \"D\": \"Green\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000328", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the image? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Blinking green\", \"D\": \"Green\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000328", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Oil slick\", \"B\": \"Slippery\", \"C\": \"Dry\", \"D\": \"Debris scattered\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000329", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000677.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000677.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Oil slick\", \"B\": \"Slippery\", \"C\": \"Dry\", \"D\": \"Debris scattered\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000329", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Parking lot\", \"B\": \"Intersection\", \"C\": \"Suburban\", \"D\": \"Bridge\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000330", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Parking lot\", \"B\": \"Intersection\", \"C\": \"Suburban\", \"D\": \"Bridge\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000330", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Dawn\", \"B\": \"Low light\", \"C\": \"Strong shadows\", \"D\": \"Under tunnel\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000331", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the lighting in this scene? You should answer one choice from four answers \"A\": \"Dawn\", \"B\": \"Low light\", \"C\": \"Strong shadows\", \"D\": \"Under tunnel\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000331", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Heavy\", \"B\": \"Rush hour\", \"C\": \"Lane merging\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000332", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Heavy\", \"B\": \"Rush hour\", \"C\": \"Lane merging\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000332", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are described in the scene? You should answer one choice from four answers \"A\": \"Stormy\", \"B\": \"Sleet\", \"C\": \"Mist\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000333", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are described in the scene? You should answer one choice from four answers \"A\": \"Stormy\", \"B\": \"Sleet\", \"C\": \"Mist\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000333", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Are there any visible traffic lights in the scene? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"Flashing red\", \"C\": \"Flashing yellow\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000334", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Are there any visible traffic lights in the scene? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"Flashing red\", \"C\": \"Flashing yellow\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000334", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Construction debris\", \"B\": \"Uneven\", \"C\": \"Rough surface\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000335", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000685.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000685.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road? You should answer one choice from four answers \"A\": \"Construction debris\", \"B\": \"Uneven\", \"C\": \"Rough surface\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000335", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Roundabout\", \"C\": \"Suburban\", \"D\": \"Urban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000336", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Roundabout\", \"C\": \"Suburban\", \"D\": \"Urban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000336", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Dusk\", \"C\": \"Blinding sunlight\", \"D\": \"Hazy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000337", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Dusk\", \"C\": \"Blinding sunlight\", \"D\": \"Hazy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000337", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Heavy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000338", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Heavy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000338", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Sandstorm\", \"C\": \"Rainy\", \"D\": \"Drizzling\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000339", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Sandstorm\", \"C\": \"Rainy\", \"D\": \"Drizzling\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000339", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Flashing red\", \"D\": \"Pedestrian signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000340", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Yellow\", \"C\": \"Flashing red\", \"D\": \"Pedestrian signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000340", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Slippery\", \"B\": \"Water puddles\", \"C\": \"Dry\", \"D\": \"Uneven\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000341", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000693.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000693.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Slippery\", \"B\": \"Water puddles\", \"C\": \"Dry\", \"D\": \"Uneven\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000341", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Suburban\", \"C\": \"Roundabout\", \"D\": \"Subway entrance\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000342", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Service station\", \"B\": \"Suburban\", \"C\": \"Roundabout\", \"D\": \"Subway entrance\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000342", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Glare\", \"B\": \"Low light\", \"C\": \"Partly cloudy\", \"D\": \"Blinding sunlight\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000343", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in the scene? You should answer one choice from four answers \"A\": \"Glare\", \"B\": \"Low light\", \"C\": \"Partly cloudy\", \"D\": \"Blinding sunlight\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000343", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Detour traffic\", \"C\": \"Light\", \"D\": \"Rush hour\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000344", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in the scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Detour traffic\", \"C\": \"Light\", \"D\": \"Rush hour\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000344", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in the scene? You should answer one choice from four answers \"A\": \"Sleet\", \"B\": \"Sunny\", \"C\": \"Snowstorm\", \"D\": \"Rainy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000345", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in the scene? You should answer one choice from four answers \"A\": \"Sleet\", \"B\": \"Sunny\", \"C\": \"Snowstorm\", \"D\": \"Rainy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000345", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Are traffic lights visible in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"Flashing yellow\", \"C\": \"None\", \"D\": \"Covered by tree\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000346", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Are traffic lights visible in the scene? You should answer one choice from four answers \"A\": \"Reflected signal\", \"B\": \"Flashing yellow\", \"C\": \"None\", \"D\": \"Covered by tree\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000346", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Snow-covered\", \"C\": \"Debris scattered\", \"D\": \"Partially damaged\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000347", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000701.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000701.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Snow-covered\", \"C\": \"Debris scattered\", \"D\": \"Partially damaged\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000347", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Airport roadway\", \"B\": \"Shopping district\", \"C\": \"Urban\", \"D\": \"Market street\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000348", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Airport roadway\", \"B\": \"Shopping district\", \"C\": \"Urban\", \"D\": \"Market street\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000348", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dawn\", \"B\": \"Low light\", \"C\": \"Smoky\", \"D\": \"Strong shadows\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000349", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Dawn\", \"B\": \"Low light\", \"C\": \"Smoky\", \"D\": \"Strong shadows\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000349", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Heavy\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000350", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Heavy\", \"C\": \"Vehicle queue\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000350", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Overcast\", \"C\": \"Mist\", \"D\": \"Snowy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000351", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Overcast\", \"C\": \"Mist\", \"D\": \"Snowy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000351", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Flashing yellow\", \"C\": \"None\", \"D\": \"Yellow\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000352", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Flashing yellow\", \"C\": \"None\", \"D\": \"Yellow\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000352", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Slippery\", \"B\": \"Wet\", \"C\": \"Dry\", \"D\": \"Snowy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000353", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000709.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000709.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Slippery\", \"B\": \"Wet\", \"C\": \"Dry\", \"D\": \"Snowy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000353", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Rural\", \"B\": \"Harbor area\", \"C\": \"Urban\", \"D\": \"Airport roadway\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000354", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Rural\", \"B\": \"Harbor area\", \"C\": \"Urban\", \"D\": \"Airport roadway\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000354", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Strong shadows\", \"B\": \"Low light\", \"C\": \"Dawn\", \"D\": \"Rain-blurred\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000355", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Strong shadows\", \"B\": \"Low light\", \"C\": \"Dawn\", \"D\": \"Rain-blurred\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000355", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Rush hour\", \"C\": \"Lane merging\", \"D\": \"Moderate\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000356", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic flow in the scene? You should answer one choice from four answers \"A\": \"Light\", \"B\": \"Rush hour\", \"C\": \"Lane merging\", \"D\": \"Moderate\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000356", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are evident in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Foggy\", \"C\": \"Stormy\", \"D\": \"Dusty\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000357", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are evident in the scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Foggy\", \"C\": \"Stormy\", \"D\": \"Dusty\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000357", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"Reflected signal\", \"C\": \"Flashing yellow\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000358", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Can you determine the status of the traffic lights in the scene? You should answer one choice from four answers \"A\": \"Off / malfunctioning\", \"B\": \"Reflected signal\", \"C\": \"Flashing yellow\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000358", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Gravel\", \"B\": \"Leaf-covered\", \"C\": \"Bumpy\", \"D\": \"Dry\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000359", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000717.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000717.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface in the scene? You should answer one choice from four answers \"A\": \"Gravel\", \"B\": \"Leaf-covered\", \"C\": \"Bumpy\", \"D\": \"Dry\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000359", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Campus area\", \"B\": \"Rural\", \"C\": \"Suburban\", \"D\": \"Construction zone\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000360", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Campus area\", \"B\": \"Rural\", \"C\": \"Suburban\", \"D\": \"Construction zone\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000360", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Dim streetlight visibility\", \"C\": \"Low light\", \"D\": \"Dawn\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000361", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Reflected glare\", \"B\": \"Dim streetlight visibility\", \"C\": \"Low light\", \"D\": \"Dawn\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000361", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Lane merging\", \"C\": \"Light\", \"D\": \"Vehicle queue\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000362", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Lane merging\", \"C\": \"Light\", \"D\": \"Vehicle queue\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000362", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather condition is suggested by the scene? You should answer one choice from four answers \"A\": \"Cloudy\", \"B\": \"Drizzling\", \"C\": \"Sleet\", \"D\": \"Sunny\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000363", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather condition is suggested by the scene? You should answer one choice from four answers \"A\": \"Cloudy\", \"B\": \"Drizzling\", \"C\": \"Sleet\", \"D\": \"Sunny\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000363", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Dim / unclear\", \"C\": \"Red\", \"D\": \"Countdown timer visible\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000364", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is the traffic light visible in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Dim / unclear\", \"C\": \"Red\", \"D\": \"Countdown timer visible\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000364", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Muddy\", \"B\": \"Construction debris\", \"C\": \"Dry\", \"D\": \"Blocked by obstacle\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000365", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000725.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000725.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Muddy\", \"B\": \"Construction debris\", \"C\": \"Dry\", \"D\": \"Blocked by obstacle\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000365", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Airport roadway\", \"C\": \"Highway\", \"D\": \"Forest road\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000366", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Airport roadway\", \"C\": \"Highway\", \"D\": \"Forest road\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000366", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Strong shadows\", \"B\": \"Under tunnel\", \"C\": \"Flashing lights\", \"D\": \"Low light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000367", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Strong shadows\", \"B\": \"Under tunnel\", \"C\": \"Flashing lights\", \"D\": \"Low light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000367", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic density in the scene? You should answer one choice from four answers \"A\": \"Heavy\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Rush hour\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000368", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current traffic density in the scene? You should answer one choice from four answers \"A\": \"Heavy\", \"B\": \"Light\", \"C\": \"Moderate\", \"D\": \"Rush hour\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000368", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are evident in the scene? You should answer one choice from four answers \"A\": \"Hailing\", \"B\": \"Sunny\", \"C\": \"Snowstorm\", \"D\": \"Cloudy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000369", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are evident in the scene? You should answer one choice from four answers \"A\": \"Hailing\", \"B\": \"Sunny\", \"C\": \"Snowstorm\", \"D\": \"Cloudy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000369", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic light ahead? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Flashing red\", \"C\": \"Blinking green\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000370", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the status of the traffic light ahead? You should answer one choice from four answers \"A\": \"Red\", \"B\": \"Flashing red\", \"C\": \"Blinking green\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000370", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Blocked by obstacle\", \"B\": \"Dry\", \"C\": \"Muddy\", \"D\": \"Worn markings\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000371", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000733.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000733.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Blocked by obstacle\", \"B\": \"Dry\", \"C\": \"Muddy\", \"D\": \"Worn markings\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000371", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Construction zone\", \"C\": \"Campus area\", \"D\": \"Intersection\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000372", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What environment does this scene represent? You should answer one choice from four answers \"A\": \"Urban\", \"B\": \"Construction zone\", \"C\": \"Campus area\", \"D\": \"Intersection\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000372", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Partly cloudy\", \"B\": \"Low light\", \"C\": \"Overexposed\", \"D\": \"Reflected glare\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000373", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Partly cloudy\", \"B\": \"Low light\", \"C\": \"Overexposed\", \"D\": \"Reflected glare\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000373", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Congested\", \"C\": \"Rush hour\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000374", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Accident blockage\", \"B\": \"Congested\", \"C\": \"Rush hour\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000374", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Smoggy\", \"C\": \"Mist\", \"D\": \"Stormy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000375", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What are the weather conditions in this scene? You should answer one choice from four answers \"A\": \"Sunny\", \"B\": \"Smoggy\", \"C\": \"Mist\", \"D\": \"Stormy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000375", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Green\", \"B\": \"None\", \"C\": \"Off / malfunctioning\", \"D\": \"Dim / unclear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000376", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"Green\", \"B\": \"None\", \"C\": \"Off / malfunctioning\", \"D\": \"Dim / unclear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000376", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Snowy\", \"C\": \"Construction debris\", \"D\": \"Muddy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000377", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000741.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000741.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the surface state of the road in this scene? You should answer one choice from four answers \"A\": \"Dry\", \"B\": \"Snowy\", \"C\": \"Construction debris\", \"D\": \"Muddy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000377", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Market street\", \"B\": \"Construction zone\", \"C\": \"Intersection\", \"D\": \"Urban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000378", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Market street\", \"B\": \"Construction zone\", \"C\": \"Intersection\", \"D\": \"Urban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Urban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000378", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Glare\", \"C\": \"Flashing lights\", \"D\": \"Clear\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000379", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Low light\", \"B\": \"Glare\", \"C\": \"Flashing lights\", \"D\": \"Clear\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000379", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Rush hour\", \"C\": \"Light\", \"D\": \"Vehicle queue\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000380", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Rush hour\", \"C\": \"Light\", \"D\": \"Vehicle queue\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000380", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather condition is suggested by the scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Mist\", \"C\": \"Sunny\", \"D\": \"Flurries\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "C" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000381", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather condition is suggested by the scene? You should answer one choice from four answers \"A\": \"Sandstorm\", \"B\": \"Mist\", \"C\": \"Sunny\", \"D\": \"Flurries\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000381", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Countdown timer visible\", \"C\": \"Off / malfunctioning\", \"D\": \"Reflected signal\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "A" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000382", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. Is there a visible traffic light in the scene? You should answer one choice from four answers \"A\": \"None\", \"B\": \"Countdown timer visible\", \"C\": \"Off / malfunctioning\", \"D\": \"Reflected signal\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000382", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Slippery\", \"B\": \"Dry\", \"C\": \"Blocked by obstacle\", \"D\": \"Cracked asphalt\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000383", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000749.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000749.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Slippery\", \"B\": \"Dry\", \"C\": \"Blocked by obstacle\", \"D\": \"Cracked asphalt\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000383", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Market street\", \"B\": \"Campus area\", \"C\": \"Airport roadway\", \"D\": \"Suburban\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000384", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Scene type", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What type of environment is depicted in the scene? You should answer one choice from four answers \"A\": \"Market street\", \"B\": \"Campus area\", \"C\": \"Airport roadway\", \"D\": \"Suburban\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Suburban" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000384", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Smoky\", \"B\": \"Low light\", \"C\": \"Rain-blurred\", \"D\": \"Foggy\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000385", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Visibility", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. How is the visibility in this scene? You should answer one choice from four answers \"A\": \"Smoky\", \"B\": \"Low light\", \"C\": \"Rain-blurred\", \"D\": \"Foggy\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Low light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000385", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Congested\", \"C\": \"Accident blockage\", \"D\": \"Light\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000386", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Traffic flow", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the traffic density in this scene? You should answer one choice from four answers \"A\": \"Moderate\", \"B\": \"Congested\", \"C\": \"Accident blockage\", \"D\": \"Light\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Light" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000386", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Dusty\", \"B\": \"Sunny\", \"C\": \"Stormy\", \"D\": \"Snowstorm\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000387", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Weather", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What weather conditions are present in the scene? You should answer one choice from four answers \"A\": \"Dusty\", \"B\": \"Sunny\", \"C\": \"Stormy\", \"D\": \"Snowstorm\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Sunny" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000387", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current state of the traffic lights ahead? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Blinking green\", \"C\": \"Temporary road sign light\", \"D\": \"None\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "D" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000388", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Traffic light", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the current state of the traffic lights ahead? You should answer one choice from four answers \"A\": \"Covered by tree\", \"B\": \"Blinking green\", \"C\": \"Temporary road sign light\", \"D\": \"None\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "None" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000388", "subtask": "label_text" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Sandy\", \"B\": \"Dry\", \"C\": \"Icy\", \"D\": \"Rough surface\". Respond with ONLY the letter of the selected option: A, B, C, or D. Do not include the option text, punctuation, or explanation. For example, respond exactly as 'A'." }, { "from": "gpt", "value": "B" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000389", "subtask": "option_letter" }, { "image": "data/perception/dsec/test/zurich_city_12_a/image/000757.png", "event": "data/perception/dsec/test/zurich_city_12_a/event/000757.npz", "category": "Road condition", "conversations": [ { "from": "human", "value": "Suppose you are driving, and I'm providing you with the image captured by the car's front center. What is the condition of the road surface? You should answer one choice from four answers \"A\": \"Sandy\", \"B\": \"Dry\", \"C\": \"Icy\", \"D\": \"Rough surface\". Respond with ONLY the corresponding option text exactly as written. Do not include the option letter or any explanation. For example, respond exactly as 'Urban'." }, { "from": "gpt", "value": "Dry" } ], "original_id": "perception/dsec/dsec_test_perception_hard.json:000389", "subtask": "label_text" } ]