conversations listlengths 2 6 | images images listlengths 1 1 | data_path stringclasses 1
value | category stringclasses 1
value | subset stringclasses 13
values | source_file stringclasses 958
values | source_row_group int32 0 46 | source_row int32 0 99 | relevance_min int8 3 5 | visual_dependency_min int8 3 5 | image_correspondence_min int8 3 5 | formatting_min int8 3 5 | image_sha256 listlengths 1 1 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
[
{
"from": "human",
"value": "<image>\nProvide a one-sentence caption for the provided image."
},
{
"from": "gpt",
"value": "Player number 42 getting ready to swing at a pitch to try to hit a baseball."
}
] | Captioning & Knowledge | textcaps | textcaps/train-00035-of-00037.parquet | 5 | 56 | 4 | 5 | 5 | 4 | [
"6c3137bd85a4f60a5d7cc0528c2fb4913b1a73eef60db47e751428e611bb99df"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you elaborate on the elements of the picture provided?"
},
{
"from": "gpt",
"value": "In the heart of the image, a wooden table sets the stage for a casual dining scene. Dominating the view are two white plates, their pristine surfaces contrasting with t... | Captioning & Knowledge | sharegpt4v(coco) | sharegpt4v(coco)/train-00022-of-00041.parquet | 10 | 1 | 4 | 5 | 3 | 4 | [
"7b5853d0a70d97e6e6c1c60e8ea45c870dd53d9bcea3cc26f3428360ac2af13c"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease describe the situation in the picture as clearly as possible."
},
{
"from": "gpt",
"value": "The image shows the facade of a building with traditional Chinese architectural elements. The most prominent feature in the image is a large signboard at the ... | Captioning & Knowledge | sharegpt4o | sharegpt4o/train-00008-of-00080.parquet | 4 | 54 | 5 | 5 | 3 | 5 | [
"4b394417dde7c73d67dde96adfde04bca97d087b608fefc6ee8aed1cb119ca37"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease provide a concise description of this image."
},
{
"from": "gpt",
"value": "In this picture we can see a man wore a helmet and riding a bicycle on the ground. In the background we can see trees."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00028-of-00043.parquet | 26 | 75 | 3 | 5 | 5 | 4 | [
"4f1d8d5f1db8e0dd42f655df181cdc0de70907e609d39dc1904b14ffc2e5c73d"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease describe what you see in this image in detail."
},
{
"from": "gpt",
"value": "The image displays a section of text on a beige background. The text is in English and is formatted in a standard article layout, with a title at the top and the main body o... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00097-of-00293.parquet | 27 | 78 | 5 | 5 | 4 | 5 | [
"ce4c5c1711dcc5b2bde4f9d1c9054cce16c13a15c8d593fde6285821f9e16ad4"
] | ||
[
{
"from": "human",
"value": "<image>\n\nWhat landmark is shown in the image?"
},
{
"from": "gpt",
"value": "This appears to be related to Wallfahrtskirche Heilig Kreuz (Westerndorf am Wasen)."
},
{
"from": "human",
"value": "Where can this place be found?"
},
{
"from": "gpt",... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00313-of-00379.parquet | 5 | 40 | 5 | 4 | 4 | 4 | [
"b533c3d9d4983543edab19b38ff90e37bf7258aec1aec0f9d6d656599479fecc"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease provide a comprehensive description of this image."
},
{
"from": "gpt",
"value": "The image displays two rattan chairs with a black and white color scheme. The chairs are positioned side by side on a neutral-toned floor against a plain, light-colored ... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00196-of-00293.parquet | 5 | 70 | 5 | 5 | 5 | 4 | [
"f3e0974371cbc8b2a251056447c509a871873fe6d6d3ce3eb924f6d71a706cf9"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat items or people are prominent in the picture?"
},
{
"from": "gpt",
"value": "The image depicts a hockey arena filled with a large audience, evident from the densely packed seating areas visible in the foreground and background. The arena is dome-shaped ... | Captioning & Knowledge | sharegpt4o | sharegpt4o/train-00054-of-00080.parquet | 4 | 65 | 5 | 5 | 3 | 5 | [
"6f3dd4ab0e499d56cc0b64064fbdaaba69e49c286add403a7470b74263ea2063"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat do you observe in this photograph?"
},
{
"from": "gpt",
"value": "The image displays a graphic with text predominantly in the center against a white background. The text is arranged in a hierarchical manner with varying font sizes and colors. The larges... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00175-of-00293.parquet | 7 | 63 | 5 | 5 | 3 | 5 | [
"f75391f709c8a4de115c45d2f1565b334f22d02cf9c2a1cea341df847e3d398f"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe this image in one or two sentences."
},
{
"from": "gpt",
"value": "There are many persons standing and sitting. There is a tent. And there is a sign board. There is a table and chair. A lady is sitting on chair. There are some vehicles. On the text ... | Captioning & Knowledge | localized_narratives | localized_narratives/train-00013-of-00043.parquet | 11 | 53 | 4 | 5 | 3 | 4 | [
"b2a72cd4b7e7a87cf41401d42cebb32d68dea6b6143bf8955c422b7f5e035de6"
] | ||
[
{
"from": "human",
"value": "<image>\nExplain the visual content of the image in great detail."
},
{
"from": "gpt",
"value": "The scene features a group of children gathered on a sidewalk, with some of them standing near the street. The children are engaged in various activities, such as eating ... | Captioning & Knowledge | image_textualization(filtered) | image_textualization(filtered)/train-00019-of-00080.parquet | 1 | 92 | 5 | 5 | 3 | 4 | [
"4dd1797346f4fca4c79422b51b05abbd0b61a53abcd3e9a8d04d977681deb398"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe this image in one or two sentences."
},
{
"from": "gpt",
"value": "As we can see fire caught in the forest and many people are seeing that fire"
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00008-of-00043.parquet | 39 | 46 | 3 | 5 | 4 | 4 | [
"9e2dda06d46b069b451373d54d500e6dcf2c5943a266b7c439921cc2ccb45a47"
] | ||
[
{
"from": "human",
"value": "<image>\nProvide a one-sentence caption for the provided image."
},
{
"from": "gpt",
"value": "A crowd of players is right near the little green for the fire exit."
}
] | Captioning & Knowledge | textcaps | textcaps/train-00013-of-00037.parquet | 3 | 10 | 3 | 5 | 4 | 4 | [
"3decabaea29bfc14a6aedfe588b9b57db3bc46ff320b6a462ee95818377f2165"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see people in black, white, darkgray, gray, and maroon tones, people in black, white, gray, and maroon tones, people in black, gray, lightgray, and darkgray to... | Captioning & Knowledge | coco_colors | coco_colors/train-00056-of-00111.parquet | 4 | 88 | 3 | 5 | 3 | 4 | [
"370c37f2458b9a85f707282c569eca0b2fa000ebc66bdbf2d8f2177affcf220b"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you elaborate on the elements of the picture provided?"
},
{
"from": "gpt",
"value": "In the center of the image, a female athlete is captured in mid-air, her body arched back slightly as she jumps with her arms raised high above her head. She's wearing ... | Captioning & Knowledge | sharegpt4v(llava) | sharegpt4v(llava)/train-00011-of-00012.parquet | 16 | 4 | 5 | 5 | 3 | 4 | [
"e980e8ae8410e6f0585e0c887dc5c0213a9de3fb334cdbab822a3b68a1540190"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the scene and objects visible in this image."
},
{
"from": "gpt",
"value": "The image displays a vibrant graphic with a central circular motif that resembles an astrological diagram, commonly associated with the zodiac signs. This circular design is... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00194-of-00293.parquet | 0 | 87 | 4 | 5 | 3 | 4 | [
"b4c5632a2056819c829640ac2b0bcbe95b0567409d64fdf6183bef7686ceafb3"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you elaborate on the elements of the picture provided?"
},
{
"from": "gpt",
"value": "The image captures a vibrant scene of a large gathering of people in what appears to be a festive, parade-like setting. The crowd is a sea of color, with individuals ho... | Captioning & Knowledge | sharegpt4v(sam) | sharegpt4v(sam)/train-00027-of-00064.parquet | 0 | 73 | 4 | 5 | 3 | 4 | [
"f9498e070658c3a58c2b7693b0f5af478fa4a89bf4a51ebaf1790fbf4382b5c9"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you describe the main features of this image for me?"
},
{
"from": "gpt",
"value": "In the image, there is a male tennis player taking a break on a tennis court. He is sitting on a white bench, which is located on the green surface of the court. The play... | Captioning & Knowledge | sharegpt4v(coco) | sharegpt4v(coco)/train-00013-of-00041.parquet | 0 | 15 | 5 | 5 | 3 | 4 | [
"f330683eb4c351710def1ac87a8124f077e5369f5ac429bb2e1b12ee99c8a25d"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you describe the main features of this image for me?"
},
{
"from": "gpt",
"value": "In the heart of a bustling city, an unusual sight unfolds. An orange flatbed truck, emblazoned with the words \"CONVOI EXCEPTIONNEL\", is transporting a unique cargo. On ... | Captioning & Knowledge | image_textualization(filtered) | image_textualization(filtered)/train-00062-of-00080.parquet | 2 | 35 | 4 | 5 | 3 | 4 | [
"bed0a72e6d77753bdca4302f1966093bb0cc8d51637f96deb794186ab6aec9e2"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the visual elements and composition of this image."
},
{
"from": "gpt",
"value": "This image features a bright yellow toddler-sized premium T-shirt with a graphic design printed on the front. The design consists of a black and white Yin-Yang symbol ... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00227-of-00293.parquet | 24 | 54 | 5 | 5 | 3 | 5 | [
"26baa14ad48e42b1950f4df340db8ac68bcc45058ef70c81bf98d60738f948bc"
] | ||
[
{
"from": "human",
"value": "<image>\nAnalyze the image in a comprehensive and detailed manner."
},
{
"from": "gpt",
"value": "In the heart of a tropical forest, a banana tree stands tall, its large green leaves reaching out towards the sky. The leaves, slightly wilted and speckled with a few br... | Captioning & Knowledge | sharegpt4v(coco) | sharegpt4v(coco)/train-00016-of-00041.parquet | 1 | 36 | 3 | 5 | 3 | 5 | [
"744ec2548ce440d25ebb8adc0e809488fa1a0898dc4c89d047edd2d6393c521a"
] | ||
[
{
"from": "human",
"value": "<image>\nProvide a detailed description of the image."
},
{
"from": "gpt",
"value": "The image features a storefront that appears to be part of a larger building with intricate wooden designs at the top. The notable components of this image include:\n\n1. **Shop Sign... | Captioning & Knowledge | sharegpt4o | sharegpt4o/train-00058-of-00080.parquet | 2 | 79 | 5 | 5 | 3 | 5 | [
"4051c5abb2caaa482df963cb9c5152493c9e18c13769630668d23a197290ed52"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease provide a concise description of this image."
},
{
"from": "gpt",
"value": "This is black and white image, we can see a few people lying on the bed, and we can see the dark background."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00025-of-00043.parquet | 36 | 83 | 3 | 5 | 3 | 4 | [
"97b2b73d50409ae30ded1ab638d4832756470cd7de5b33f3bd2ba4a8dcabb033"
] | ||
[
{
"from": "human",
"value": "<image>\n\nDescribe the landmark in the photo."
},
{
"from": "gpt",
"value": "This is São Jorge Castle."
},
{
"from": "human",
"value": "What is the location of this landmark?"
},
{
"from": "gpt",
"value": "It is located in Lisbon, Portugal."
... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00327-of-00379.parquet | 3 | 50 | 4 | 5 | 3 | 4 | [
"311771da91175d970ff57bc326c2f0404189d8d49cbcbabd70d0b8c701e60c14"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat can you see and identify in this photograph?"
},
{
"from": "gpt",
"value": "The image displays a screenshot from the National Weather Service Mesonet Observations website. The top portion of the image features the logo of the National Oceanic and Atmosp... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00181-of-00293.parquet | 3 | 12 | 5 | 5 | 5 | 4 | [
"9c5e5b8c011955051e74cb7a9b82658255e758f819d1bf7bff6a1bc8f8b3ec93"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you tell me what's happening in this picture?"
},
{
"from": "gpt",
"value": "The image features a person standing with their back to the camera, wearing a bright orange shirt and a black backpack with a visible logo. The person is surrounded by a vibrant... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00177-of-00293.parquet | 7 | 46 | 5 | 5 | 3 | 5 | [
"6698f22dc7175e31438bc54bc8ff34495d80740e7948a560a8779c643d42b2ae"
] | ||
[
{
"from": "human",
"value": "<image>\nProvide a one-sentence caption for the provided image."
},
{
"from": "gpt",
"value": "Bookstall in white and blue inside a station."
}
] | Captioning & Knowledge | textcaps | textcaps/train-00020-of-00037.parquet | 5 | 60 | 5 | 5 | 4 | 4 | [
"3939ca0ebd8c7b468196430a19884dfcaf34af44c6c7429e445130d20afb94f2"
] | ||
[
{
"from": "human",
"value": "<image>\n\nCan you identify the place in the picture?"
},
{
"from": "gpt",
"value": "This is Бережанський замок."
},
{
"from": "human",
"value": "What is the location of this landmark?"
},
{
"from": "gpt",
"value": "It is located in Berezhany,... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00012-of-00379.parquet | 6 | 49 | 3 | 4 | 3 | 4 | [
"90747adbe630f1e4d5045bdc189102b55bf097bdc144130856584075cfb210dc"
] | ||
[
{
"from": "human",
"value": "<image>\nIn one or two sentences, can you explain what this image depicts?"
},
{
"from": "gpt",
"value": "In this picture there are few eatables placed on a stand and there are few other eatables placed on either sides of it and there are two persons standing behind ... | Captioning & Knowledge | localized_narratives | localized_narratives/train-00016-of-00043.parquet | 44 | 9 | 3 | 5 | 3 | 4 | [
"32018949384073d010d968e163e1be404a59ab7fad34a2868e1b1648c4d0a612"
] | ||
[
{
"from": "human",
"value": "<image>\nProvide a one-sentence caption for the provided image."
},
{
"from": "gpt",
"value": "An LG smartphone is displayed surrounded by a white background."
}
] | Captioning & Knowledge | textcaps | textcaps/train-00003-of-00037.parquet | 4 | 61 | 5 | 5 | 3 | 4 | [
"1c7b779d942732cba6767050f698f01362bb0b3ed757194dbbd35302781f4eca"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the visual elements and composition of this image."
},
{
"from": "gpt",
"value": "This is an image of a book cover with a graphical and textual design. The background of the cover is white. The title of the book \"Leadership Basics for Frontline Man... | Captioning & Knowledge | laion_gpt4v | laion_gpt4v/train-00002-of-00007.parquet | 12 | 94 | 5 | 5 | 3 | 4 | [
"2ad3a963f776c7f96ad12217391cd0348901055ed362a7aad8c7baf81cbd71fc"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you elaborate on the elements of the picture provided?"
},
{
"from": "gpt",
"value": "In the center of the image, a cozy living room unfolds. The room is bathed in a warm beige color, from the walls to the carpet, creating a harmonious and inviting atmos... | Captioning & Knowledge | image_textualization(filtered) | image_textualization(filtered)/train-00070-of-00080.parquet | 4 | 43 | 5 | 5 | 3 | 4 | [
"663a2037dda69b13e2f48f104095fbb999790ffaeb705c92e180627865f5aba6"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat can you see and identify in this photograph?"
},
{
"from": "gpt",
"value": "The image displays a baby wearing a white quilted Chanel handbag. The handbag features the iconic Chanel logo in gold lettering on a white background. The baby is also wearing p... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00129-of-00293.parquet | 12 | 80 | 5 | 5 | 3 | 5 | [
"a35257b931c216123df5619c01af815d81f2211172df927185ce238363312a39"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you tell me what's happening in this picture?"
},
{
"from": "gpt",
"value": "The image features a stylized illustration of a clouded leopard, which is a feline species, depicted in a dynamic pose as if it is running. The leopard is anthropomorphized with... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00245-of-00293.parquet | 12 | 81 | 5 | 5 | 3 | 5 | [
"5cab04ec58a5394b37e607473d1f49b5229412500ae1433cf40d7727d5fcea6e"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe this image in one or two sentences."
},
{
"from": "gpt",
"value": "In this image there is a wall, for that wall there frame, in that frame there is a letter and glove."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00003-of-00043.parquet | 37 | 89 | 3 | 5 | 3 | 4 | [
"c96b1cd092b52a4c5a6f7a4ca85a25b57133970ed0abc23c0dd416fad2535626"
] | ||
[
{
"from": "human",
"value": "<image>\nCould you give a brief overview of what you see in this image?"
},
{
"from": "gpt",
"value": "In this image I see white plates on which there is food which is of brown, red and green in color and I see the maroon color thing over here."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00029-of-00043.parquet | 8 | 16 | 3 | 5 | 3 | 4 | [
"3b56d141864c8dfb651810ac06350e0c22e8a8f1fc3f06e2f6a7ef54bf6b6439"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see bowl in gray, darkgray, lightgray, and black tones and spoon in gray, white, darkgray, and lightgray tones in this image."
}
] | Captioning & Knowledge | coco_colors | coco_colors/train-00000-of-00111.parquet | 5 | 53 | 4 | 5 | 3 | 4 | [
"328bdb17af962afc742ea176f4663b4e0d88c466f0591973d3164fedcf50a480"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat details are worth noting in the picture?"
},
{
"from": "gpt",
"value": "The image depicts a man and a woman posing together in what appears to be an indoor setting, possibly a restaurant. The man is on the left side of the image, and the woman is on the... | Captioning & Knowledge | sharegpt4o | sharegpt4o/train-00041-of-00080.parquet | 1 | 48 | 5 | 5 | 3 | 4 | [
"961a951b76acdab495ba7f8f78703e161d48d1975690cc48c083edcb8eb01f10"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat are the key features and details in this picture?"
},
{
"from": "gpt",
"value": "The image depicts a warm and cheerful family scene within a bedroom setting. There are three individuals: a man, a woman, and a young boy. The man, wearing a white T-shirt,... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00071-of-00293.parquet | 15 | 8 | 4 | 5 | 3 | 4 | [
"e58bbeb7acf70d6d26a9de077da3a4303b763ea6b92c0c796ab25744cf457fb7"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat is this photo about'?"
},
{
"from": "gpt",
"value": "The image presents an aerial view of a cityscape, with a significant power plant standing out in the background. The power plant is characterized by two tall smokestacks, painted in alternating red an... | Captioning & Knowledge | sharegpt4v(sam) | sharegpt4v(sam)/train-00033-of-00064.parquet | 1 | 15 | 5 | 5 | 3 | 4 | [
"c00880d9f0c127ba2a22609c4e759c2242f770e3ba11ed789ccc59c7425ead6c"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you explain in detail what is happening in the picture?"
},
{
"from": "gpt",
"value": "The image shows an expansive natural landscape that features a rocky formation of distinct jagged peaks and rounded boulders. The rocks show signs of erosion, suggesti... | Captioning & Knowledge | sharegpt4o | sharegpt4o/train-00029-of-00080.parquet | 3 | 72 | 5 | 5 | 3 | 5 | [
"d42d6d3c35512098425e3ee3130ca95d64375d40d03214bb9577a182a83719a2"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the scene and objects visible in this image."
},
{
"from": "gpt",
"value": "The image displays a variety of personal care products on sale, likely within a store. In the foreground, there is a package of \"Mach 3 Turbo\" razor blades, priced at $15.... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00242-of-00293.parquet | 27 | 62 | 4 | 5 | 3 | 4 | [
"613981ae225b944d954c2aea17894d1d5a81a4df19a5759bc3118556e3f2d6f3"
] | ||
[
{
"from": "human",
"value": "<image>\n\nWhat landmark is shown in the image?"
},
{
"from": "gpt",
"value": "This is Hôtel du Parlement."
},
{
"from": "human",
"value": "Can you tell me the location?"
},
{
"from": "gpt",
"value": "It is located in Quebec, Quebec, Canada."
... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00134-of-00379.parquet | 4 | 49 | 5 | 3 | 3 | 4 | [
"2a32ff2c2b99afbdbe88b8341dbf2c4b6d3cf93fb720bc6fccbf202bc5d9e420"
] | ||
[
{
"from": "human",
"value": "<image>\nAnalyze the image in a comprehensive and detailed manner."
},
{
"from": "gpt",
"value": "The image captures a charming street-side café nestled in a bustling European city. The perspective of the photo is as if you're seated at one of the wooden tables, invi... | Captioning & Knowledge | sharegpt4v(sam) | sharegpt4v(sam)/train-00000-of-00064.parquet | 0 | 11 | 4 | 5 | 3 | 4 | [
"ba813d42db7c46a5ec3296143ad3c3d6b6f751396b7f6f9c7b2e96dd87549c0c"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe this image in one or two sentences."
},
{
"from": "gpt",
"value": "In this image we can see a plate with food item. And we can see fingers of a person holding the plate. Also there are glasses. And it is looking blur in the background."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00003-of-00043.parquet | 32 | 28 | 3 | 5 | 3 | 4 | [
"40cc1863921a942e11c26e4bd98aa0e80435c0d3798dff7e474fc9f546de4209"
] | ||
[
{
"from": "human",
"value": "<image>\n\nDescribe the landmark in the photo."
},
{
"from": "gpt",
"value": "This appears to be related to Temple of Apollo (Pompeii)."
},
{
"from": "human",
"value": "What is the location of this landmark?"
},
{
"from": "gpt",
"value": "It i... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00016-of-00379.parquet | 6 | 27 | 4 | 5 | 3 | 4 | [
"ecb2a5127c236043ece411036f1d71716bb0a6dabb0fa5778ba014621f18a482"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease provide a concise description of this image."
},
{
"from": "gpt",
"value": "It is a sports room, a person who is wearing black shirt in the front is playing table tennis ,there are also some other people behind him some of them are sitting and some of... | Captioning & Knowledge | localized_narratives | localized_narratives/train-00024-of-00043.parquet | 42 | 70 | 4 | 5 | 5 | 4 | [
"c10561b9bf826fdcf8faad025a04cb6cd563d69d4d93c6438b996cdfdc9b34b7"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the following image."
},
{
"from": "gpt",
"value": "The image features a group of people gathered around a table, with a man in a black shirt and white name tag, standing very close to the camera, pointing at a video game console. He is likely expla... | Captioning & Knowledge | image_textualization(filtered) | image_textualization(filtered)/train-00012-of-00080.parquet | 11 | 27 | 4 | 5 | 3 | 4 | [
"05e1460949c6f24053a20c7d416d70796916e687a2b7bba76b36859c016e348f"
] | ||
[
{
"from": "human",
"value": "<image>\n\nCan you identify the place in the picture?"
},
{
"from": "gpt",
"value": "This is Infusões bar."
},
{
"from": "human",
"value": "Where can this place be found?"
},
{
"from": "gpt",
"value": "It is located in Guimarães, Portugal."
... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00080-of-00379.parquet | 1 | 5 | 5 | 4 | 3 | 4 | [
"4ac0960770a2a1090b9165eff02bb448fad51472f4504fca6015b447a58dddb6"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe this image in one or two sentences."
},
{
"from": "gpt",
"value": "In the picture there are two different images collage together, there are two toys in the image both the toys are of similar type."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00002-of-00043.parquet | 13 | 77 | 3 | 5 | 3 | 4 | [
"3b7b2f94f203ec0a45ed0f0bd34f0235b5853cf450e3b34d2f4c373167912bb1"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat can you see and identify in this photograph?"
},
{
"from": "gpt",
"value": "This image depicts a kitchen countertop with various appliances and utensils. On the left, there is a stainless steel sink with a silver faucet. Above the sink, a black soap dis... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00161-of-00293.parquet | 31 | 3 | 5 | 5 | 3 | 5 | [
"a2f5dc17eb067ffda8aa955a7d4647d498e59c9f017bcc9dd11094859637c81e"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease describe the content of the picture."
},
{
"from": "gpt",
"value": "The image features a woman dressed in traditional attire, standing in an outdoor setting. Her outfit, brimming with intricate and colorful details, is rich in cultural significance.\n... | Captioning & Knowledge | sharegpt4o | sharegpt4o/train-00034-of-00080.parquet | 1 | 13 | 5 | 5 | 3 | 5 | [
"d3c14e7cf045f6ab385a920ee9d5388f48185888188d00cecec326280aaa0e15"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat is the main content of this image?"
},
{
"from": "gpt",
"value": "This image features a red Kids T-Shirt laid out flat. On the T-Shirt, there is a colorful print of a painting. The painting depicts a caricatured figure with exaggerated facial features, ... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00201-of-00293.parquet | 3 | 33 | 5 | 5 | 3 | 4 | [
"29c40273b03f7490aa0ab19b163bae2fb7828f881ac1de303872ca1af8b32beb"
] | ||
[
{
"from": "human",
"value": "<image>\n\nDescribe the landmark in the photo."
},
{
"from": "gpt",
"value": "This is St. Jakobus und St. Josef."
},
{
"from": "human",
"value": "Where can this place be found?"
},
{
"from": "gpt",
"value": "It is located in Vienna, Austria."
... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00095-of-00379.parquet | 7 | 62 | 4 | 4 | 3 | 4 | [
"aebe49b900733a41271b5c3c7604eaf18f051c958b1b66618dbe9917759d427c"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see people in black, lightgray, darkgray, tan, and gray tones, dining table in black, lightgray, darkgray, gray, and maroon tones, cake in black, lightgray, da... | Captioning & Knowledge | coco_colors | coco_colors/train-00065-of-00111.parquet | 2 | 48 | 4 | 5 | 5 | 4 | [
"1f6b1557d3aac3c385cebfda42971375598366446a6481ff47d96878b345bb0f"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you tell me what's happening in this picture?"
},
{
"from": "gpt",
"value": "The image displays a graphic with a Christmas theme, featuring a poem titled \"Adam lay ybounden.\" The text is arranged in a stanza format, with the poem's title at the top in ... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00156-of-00293.parquet | 22 | 23 | 5 | 5 | 3 | 5 | [
"d805fa64c757396fb6e18121cbcddfcec5ea99595fe727d7cf99458f04d67d8a"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see various objects in this image with different colors."
}
] | Captioning & Knowledge | coco_colors | coco_colors/train-00083-of-00111.parquet | 1 | 95 | 3 | 5 | 3 | 4 | [
"3afe025c17c67522ccfaad98066053f704ca53180517e905c3d233f83f37f886"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you elaborate on the elements of the picture provided?"
},
{
"from": "gpt",
"value": "The image features a man standing on a tennis court, ready to take a swing with a tennis racket. He is wearing a white t-shirt and black shorts, holding the racket in a... | Captioning & Knowledge | image_textualization(filtered) | image_textualization(filtered)/train-00004-of-00080.parquet | 10 | 61 | 4 | 5 | 3 | 5 | [
"e5907b8bfabe4e4384505d1ddff8969da0f04e6e373a311b85f0f6ab6f17cd9e"
] | ||
[
{
"from": "human",
"value": "<image>\n\nDescribe the landmark in the photo."
},
{
"from": "gpt",
"value": "This is Casa del Marqués de Prado Alegre."
},
{
"from": "human",
"value": "Where is this located?"
},
{
"from": "gpt",
"value": "It is located in Mexico City, Mexico... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00237-of-00379.parquet | 6 | 70 | 3 | 4 | 3 | 4 | [
"c4f48cad1080ff3e031d36a0c6403b156c8aa0b4dfcb0af5603c4e1108b92910"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat is this photo about'?"
},
{
"from": "gpt",
"value": "This black and white image captures a moment of pure exhilaration and freedom. At the heart of the scene is a surfer, clad in a wetsuit, skillfully riding a wave. The surfer is crouched low on their s... | Captioning & Knowledge | sharegpt4v(coco) | sharegpt4v(coco)/train-00037-of-00041.parquet | 6 | 17 | 5 | 5 | 3 | 5 | [
"213f53cf37e0be7a13876d2403a56db6c9f340b8e28042305fcaebf3bbdc1ec4"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see bowl in maroon, tan, lightgray, and darkgray tones, dining table in maroon, olive, and orange tones, knife in maroon, black, and gray tones, fork in maroon... | Captioning & Knowledge | coco_colors | coco_colors/train-00082-of-00111.parquet | 8 | 50 | 5 | 5 | 3 | 4 | [
"721d0f311fef30b0878f34ea7736318bfc8577537f5d4141a0dabcb8108020d8"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat is the main content of this image?"
},
{
"from": "gpt",
"value": "The image displays a white Prius α hatchback positioned on a blue car lift inside a tire service shop. The vehicle is facing to the right, with its rear end elevated, exposing the rear ti... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00024-of-00293.parquet | 17 | 38 | 5 | 5 | 3 | 4 | [
"bef41a7d842b7e287a924af85ef7ae18d4302d2fae3b9d70ba16360d1e1e3b5e"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease describe what you see in this image in detail."
},
{
"from": "gpt",
"value": "The image displays a cardboard box with text printed on its side. The box appears to be in a storage or warehouse environment, indicated by the concrete wall and floor in th... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00110-of-00293.parquet | 28 | 13 | 5 | 5 | 4 | 4 | [
"4394756fa3c681919270805a75325234e807e81471b3b8ea5a2ece3978621800"
] | ||
[
{
"from": "human",
"value": "<image>\n\nCan you identify the place in the picture?"
},
{
"from": "gpt",
"value": "This appears to be related to Boim Chapel."
},
{
"from": "human",
"value": "Where is this located?"
},
{
"from": "gpt",
"value": "It is located in Lviv, Lviv ... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00203-of-00379.parquet | 2 | 25 | 4 | 4 | 3 | 4 | [
"19dec7a12cbcad168f2cc347a3f39b7ea8b01a3c6d86539bf76c214f77b28ba2"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you describe this image briefly?"
},
{
"from": "gpt",
"value": "In this picture I can see a woman holding camera, around I can see few people, plants, trees and pillories."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00039-of-00043.parquet | 23 | 1 | 3 | 5 | 3 | 4 | [
"8b447ec09638e1414d741e7aa2ab51c41d8aa420408ec55553ee70fd900446ec"
] | ||
[
{
"from": "human",
"value": "<image>\nCould you give a brief overview of what you see in this image?"
},
{
"from": "gpt",
"value": "In the picture we can see a plant in the bowl with a stand which is white in color and the flowers are violet in color and in the background we can see a grass surf... | Captioning & Knowledge | localized_narratives | localized_narratives/train-00034-of-00043.parquet | 1 | 32 | 4 | 5 | 3 | 4 | [
"5e369f05e148c9e8536325895a2935f55d3989d66544bddebe718e37b313a175"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat's happening in the scene?"
},
{
"from": "gpt",
"value": "The image shows two horses standing in the back of a large trailer or horse trailer, which is parked on the side of the road. One horse is brown and black, while the other is black with white feet... | Captioning & Knowledge | image_textualization(filtered) | image_textualization(filtered)/train-00078-of-00080.parquet | 5 | 93 | 5 | 5 | 3 | 5 | [
"c4358dd35a9cf93d27219fd76c80cc3c8ef8239321551b64b813d15308161cf5"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease describe the content of the picture."
},
{
"from": "gpt",
"value": "The image depicts a serene coastal scene featuring a small harbor or bay. The primary object in the foreground is a fishing boat, painted in a bright red color, which is situated in c... | Captioning & Knowledge | sharegpt4o | sharegpt4o/train-00061-of-00080.parquet | 2 | 28 | 5 | 5 | 3 | 5 | [
"1829bf20180c52b795fa486ea1b4b2604096c84fd9276b74ffc4578107acb27f"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat do you observe in this photograph?"
},
{
"from": "gpt",
"value": "The image features a stylized graphic with a teal background. At the center, there is a large text in white that reads \"RESEARCH THE ALTERNATIVES.\" Below the text, there is a logo consi... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00278-of-00293.parquet | 24 | 44 | 4 | 5 | 3 | 4 | [
"684d7be9e6bdc65fd96240d9cc8544a97a574849fef360ede9ab3b4c84a1361b"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease give me a detailed overview of what this image shows."
},
{
"from": "gpt",
"value": "The image displays a variety of items related to rope splicing, arranged on a surface and within an open cardboard box. The box features a prominent blue and white la... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00071-of-00293.parquet | 36 | 11 | 5 | 5 | 5 | 4 | [
"a455ab6ee200fcd8a540dd0d47b05094f0ed07457a421536fa1414cb67fe909b"
] | ||
[
{
"from": "human",
"value": "<image>\nWrite a detailed description of the given image."
},
{
"from": "gpt",
"value": "In the image, we see the actor Henry Cavill portraying his iconic character, Superman. He stands in the center of a small town street, exuding an aura of seriousness and determin... | Captioning & Knowledge | sharegpt4v(knowledge) | sharegpt4v(knowledge)/train-00003-of-00005.parquet | 1 | 23 | 5 | 5 | 5 | 4 | [
"8295d1cdea1865e162b52e0ae4d27614f6b4dc43afc2d2d01452b4060466c858"
] | ||
[
{
"from": "human",
"value": "<image>\n\nDescribe the landmark in the photo."
},
{
"from": "gpt",
"value": "This appears to be related to Catacombes de Paris."
},
{
"from": "human",
"value": "Where can this place be found?"
},
{
"from": "gpt",
"value": "It is located in Pa... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00101-of-00379.parquet | 0 | 51 | 4 | 4 | 3 | 4 | [
"0c5283e81297375807edffb1e6182be37571e5d0f3f3d45ed88595d756f245c0"
] | ||
[
{
"from": "human",
"value": "<image>\nCould you give a brief overview of what you see in this image?"
},
{
"from": "gpt",
"value": "In this picture we can see a person holding tablet(device) in his hand. We can see a few people from left to right. There are some electronic devices and a cup on t... | Captioning & Knowledge | localized_narratives | localized_narratives/train-00031-of-00043.parquet | 30 | 83 | 4 | 5 | 3 | 4 | [
"07f6bb66d3e882d5be0993e99dfaa9c2e55d25378ea0f91549c026f662201868"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease provide a concise description of this image."
},
{
"from": "gpt",
"value": "In this image we can see a lady wearing black dress."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00030-of-00043.parquet | 41 | 15 | 4 | 5 | 3 | 4 | [
"4181372044d01e884aa2e5e9253cae69ef8d8f8c2dfd79699c2bdbd8e33b77cc"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you tell me what's happening in this picture?"
},
{
"from": "gpt",
"value": "The image displays a wicker basket filled with garlic scapes, which are the curved, green stalks of garlic, resting on a dark signboard. The signboard features white text that r... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00200-of-00293.parquet | 32 | 38 | 4 | 5 | 3 | 5 | [
"2a0542d4820ecf63868b5fb21ed738777ff775582ea9bcd00464f602d0d8f14c"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see people in white, black, darkgray, brown, and navy tones, people in white, black, brown, and maroon tones, umbrella in white, maroon, brown, and black tones... | Captioning & Knowledge | coco_colors | coco_colors/train-00095-of-00111.parquet | 6 | 87 | 4 | 5 | 3 | 4 | [
"4b5dbfc53db1a808c4d09bee1e2e842c6ce05833f5c76810a20738c3090d7916"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat do you think is going on in this snapshot?"
},
{
"from": "gpt",
"value": "The image captures a bustling scene at an airport tarmac. Dominating the frame is a large airplane, its body painted in a pristine white, adorned with a blue stripe running along ... | Captioning & Knowledge | image_textualization(filtered) | image_textualization(filtered)/train-00015-of-00080.parquet | 7 | 30 | 5 | 5 | 3 | 5 | [
"7cc7c596c078bb6bf67a9717d1d06b45d6ee9e18858efa0544503fa99ac4df43"
] | ||
[
{
"from": "human",
"value": "<image>\n\nCan you identify the place in the picture?"
},
{
"from": "gpt",
"value": "This is Очите."
},
{
"from": "human",
"value": "Where can this place be found?"
},
{
"from": "gpt",
"value": "It is located in Karlukovo, Bulgaria."
},
{
... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00320-of-00379.parquet | 4 | 55 | 4 | 4 | 3 | 4 | [
"640387c24b2dc74cd03b4af91f93c2289a2188b23167b4c7a8947d6989e885d3"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat is the main content of this image?"
},
{
"from": "gpt",
"value": "The image displays a person standing against a plain, light-colored background. The individual appears to be a male with curly hair and is wearing black-framed spectacles. He is dressed i... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00077-of-00293.parquet | 34 | 60 | 5 | 5 | 3 | 5 | [
"6d7c5f322f8980391bfa81bc3cabe43af701ed953d34518043696da9596229af"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease describe specifically what this image is about."
},
{
"from": "gpt",
"value": "The image depicts a scene at an airport where a group of passengers are boarding an airplane. The airplane is a large commercial jet, specifically an American Airlines airc... | Captioning & Knowledge | sharegpt4o | sharegpt4o/train-00050-of-00080.parquet | 1 | 35 | 5 | 5 | 3 | 4 | [
"bea7fb249695d01c15329ff430f36f3ec14c75e18cef8167c035b2f3d2037965"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease give me a detailed overview of what this image shows."
},
{
"from": "gpt",
"value": "The image is a graphic with a vibrant orange background featuring a central circular cutout with a gray gradient background. Within this cutout is a photograph of a m... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00240-of-00293.parquet | 17 | 11 | 5 | 5 | 4 | 4 | [
"7328cbb01bddf7727b8721ac156a1bc5a2a8bdaed7c48f5304ab80e1e84a65aa"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat are the key elements in this picture?"
},
{
"from": "gpt",
"value": "The image presents a beautiful arrangement of wedding invitation elements. Dominating the scene is a rectangular wedding invitation card, adorned with a mesmerizing blend of purple and... | Captioning & Knowledge | sharegpt4v(llava) | sharegpt4v(llava)/train-00009-of-00012.parquet | 10 | 65 | 4 | 5 | 3 | 4 | [
"6a1b14d76c7f882ee8fd4f137f96497368f87eb907ed206ef770c6b1d423f419"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you describe this image briefly?"
},
{
"from": "gpt",
"value": "In this image, There are two persons standing and holding a dog which is in brown color, They both kissing the dog, In the background there is a blue color curtain."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00032-of-00043.parquet | 22 | 81 | 4 | 5 | 3 | 4 | [
"4bd37b1d32ad13c418dd1d48003ed50aa11fcc5da2b6d0a64b9b9179a9f5e956"
] | ||
[
{
"from": "human",
"value": "<image>\n\nWhat is this place?"
},
{
"from": "gpt",
"value": "This is Tuel Lane car park."
},
{
"from": "human",
"value": "Where can this place be found?"
},
{
"from": "gpt",
"value": "It is located in Calderdale, England, United Kingdom."
}... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00124-of-00379.parquet | 6 | 55 | 4 | 4 | 3 | 4 | [
"c84d0e310beb16921d1b780eb5fc6b888dcc5a799d4897303f80e77090fa6201"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see sink in darkgray, lightgray, and gray tones, sink in darkgray, lightgray, and gray tones, and book in darkgray, lightgray, and gray tones in this image."
... | Captioning & Knowledge | coco_colors | coco_colors/train-00095-of-00111.parquet | 4 | 6 | 3 | 5 | 3 | 4 | [
"e0b44c7e67f953580678502be21a6b0a76e4816e24fe124b8736d42d8735f9b8"
] | ||
[
{
"from": "human",
"value": "<image>\nHow would you summarize this image in a sentence or two?"
},
{
"from": "gpt",
"value": "This is a black and white picture, in this image we can see a cat and a person wearing the spectacles, also we can see some objects."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00020-of-00043.parquet | 37 | 13 | 3 | 5 | 3 | 4 | [
"9926e921f51ff7c40b78ddd1d0e00baaf430d221152edfe4031fd6d75f157b3e"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see dining table in black, khaki, tan, and maroon tones, bowl in black, khaki, and tan tones, bowl in black, tan, khaki, and brown tones, bottle in black and t... | Captioning & Knowledge | coco_colors | coco_colors/train-00059-of-00111.parquet | 3 | 88 | 5 | 5 | 3 | 4 | [
"4f41a19e604633b76ce3bee0bee82cf9a8e839efdb24092ff0a191c9e7e56aff"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease provide a concise description of this image."
},
{
"from": "gpt",
"value": "In the image in the center we can see one house,wall and roof. In the background we can see sky,clouds,mountain,cross,trees,plants and grass."
}
] | Captioning & Knowledge | localized_narratives | localized_narratives/train-00016-of-00043.parquet | 15 | 71 | 4 | 5 | 3 | 4 | [
"9951bef4ed26354df23efc489322c2a3503230d920abf5b713c52be406798506"
] | ||
[
{
"from": "human",
"value": "<image>\n\nDescribe the landmark in the photo."
},
{
"from": "gpt",
"value": "This is Caroli kyrka."
},
{
"from": "human",
"value": "Where can this place be found?"
},
{
"from": "gpt",
"value": "It is located in Malmo, Sweden."
},
{
"f... | Captioning & Knowledge | google_landmarks | google_landmarks/train-00017-of-00379.parquet | 6 | 87 | 3 | 4 | 4 | 4 | [
"7072e5652e220723bbae6e2b8cdbbb0cfd31ee2a12ba140de2e86f21aee2fb9d"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you elaborate on the elements of the picture provided?"
},
{
"from": "gpt",
"value": "The image presents a model sheet of the character Brendan from the animated film \"Brendan and the Secret of Kells\". Brendan is depicted in six different angles: front... | Captioning & Knowledge | sharegpt4v(llava) | sharegpt4v(llava)/train-00000-of-00012.parquet | 6 | 55 | 5 | 5 | 3 | 5 | [
"1110165365e92daa30adc491436f30ea837a9c0e86c5fd9a1f0a29cb8fc563d9"
] | ||
[
{
"from": "human",
"value": "<image>\nCan you tell me what's happening in this picture?"
},
{
"from": "gpt",
"value": "The image displays a graphic with a mint green background and a text overlay in varying fonts and sizes. The text is primarily in white with some words highlighted in pink. The ... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00042-of-00293.parquet | 9 | 60 | 5 | 5 | 3 | 5 | [
"57f15f1fb4ce2a74cd5aa911f28177b34227e31a532c7e61fe2a4f8c608a0de2"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat do you observe in this photograph?"
},
{
"from": "gpt",
"value": "A young boy stands centered on a wooden bridge surrounded by lush greenery. He has curly hair and is wearing a white T-shirt with the text \"EMERGING ARTIST\" printed in black letters acr... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00123-of-00293.parquet | 15 | 25 | 5 | 5 | 3 | 4 | [
"490d512986da0a5f21d80426d4293646f377ad920ee06d5120e9825355a151e3"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat do you observe in this photograph?"
},
{
"from": "gpt",
"value": "The image displays a red Suzuki Access 125 scooter parked on the side of a road, with a green and white welcome sign reading \"Resort Town of Negril The Capital of Casual\" visible in the... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00061-of-00293.parquet | 8 | 57 | 5 | 5 | 3 | 5 | [
"7b813193b463e62acd49f4cfe70e049e48303b2c20ca94fcf1b06b376d3dc200"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat do you observe in this photograph?"
},
{
"from": "gpt",
"value": "The image displays a pair of LED lights designed to resemble the wings of a butterfly. These wings are illuminated and feature intricate patterns that mimic the natural vein structure of ... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00098-of-00293.parquet | 4 | 96 | 4 | 5 | 3 | 4 | [
"9b97f72e2de3f855bf280428d8fe6fdc299feb7c91bff35ed5a237a0ebb43a74"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see a horse in gray, black, olive, and tan tones in this image."
}
] | Captioning & Knowledge | coco_colors | coco_colors/train-00035-of-00111.parquet | 6 | 56 | 4 | 5 | 3 | 4 | [
"f93eb9e8523428f9eb3c3c527722f6d44bd361008ba6c4f78da275b72239f714"
] | ||
[
{
"from": "human",
"value": "<image>\nWhat can you see and identify in this photograph?"
},
{
"from": "gpt",
"value": "The image is a collage of four separate photos, each featuring a person with sports-themed face stickers. In the top left photo, a man with dark hair is smiling, wearing black a... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00080-of-00293.parquet | 19 | 44 | 5 | 5 | 5 | 5 | [
"c3e6e3788a5ae556c0e6c8ce8603f7e0009a17bc6ec0f11a4258ee64f796b000"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the scene and objects visible in this image."
},
{
"from": "gpt",
"value": "The image displays a close-up view of a data recovery process involving a hard disk drive (HDD). The central focus is on the HDD's read/write heads, which are located at the... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00284-of-00293.parquet | 25 | 36 | 4 | 5 | 5 | 4 | [
"5ea4dad83cd01e88ac3acbab62d6c6d4dab926a2842d9f06fbf65b71655c368d"
] | ||
[
{
"from": "human",
"value": "<image>\nPlease give me a detailed overview of what this image shows."
},
{
"from": "gpt",
"value": "The image displays a promotional poster with a yellow and black color scheme. At the top, in bold black letters against a yellow background, it reads \"PRIVATE INSPEC... | Captioning & Knowledge | densefusion_1m | densefusion_1m/train-00077-of-00293.parquet | 18 | 51 | 5 | 5 | 3 | 5 | [
"4885db913cb29782dba118454fca50e7f1d70e333d1adb04d98f60edbd2fa0f1"
] | ||
[
{
"from": "human",
"value": "<image>\nDescribe the objects in this image and their specific colors."
},
{
"from": "gpt",
"value": "I can see sandwich in darkgray, olive, tan, and maroon tones and people in darkgray and gray tones in this image."
}
] | Captioning & Knowledge | coco_colors | coco_colors/train-00074-of-00111.parquet | 2 | 57 | 4 | 5 | 3 | 4 | [
"cc5f0ac470f05d8de3f22b3168dc0a1490f95ff5ae71ea23b1831757762b9bbc"
] | ||
[
{
"from": "human",
"value": "<image>\nWrite a detailed description of the given image."
},
{
"from": "gpt",
"value": "The image portrays a lively scene of two women engaged in a dance. Both women are captured mid-motion, their arms raised high and heads tilted to the side, embodying the joy and ... | Captioning & Knowledge | sharegpt4v(llava) | sharegpt4v(llava)/train-00000-of-00012.parquet | 17 | 86 | 5 | 5 | 3 | 5 | [
"111e10eb6bd54c741f203c0eb8ee36455f41b977cb07f33a62816218dcf51150"
] |
End of preview. Expand in Data Studio
mmret-finevision-1m
A 1,000,000-row, quality-filtered, deduplicated subset of HuggingFaceM4/FineVision, repacked as conversation rows with the images inline (no external image files), for joint understanding + retrieval training of Qwen2.5-VL in the mmret project.
Each subset of FineVision keeps its own license; using this data means accepting the license of every subset it was drawn from (see the FineVision card).
Selection
- Quality: all four per-sample minimum ratings of FineVision (relevance, visual dependency, image correspondence, formatting) >= 3. Text-only subsets carry no ratings and are kept; an image row with a missing rating is excluded.
- Length: at most 4 images and an estimated
576 * images + chars / 3.5 + 64 <= 4608tokens (the 576-visual-token training budget). - Subsets with a pass rate below 5% are excluded (22 subsets, e.g. DoclingMatix, dvqa, figureqa, plotqa, mapqa, synthdog, lnqa, several CoSyn-400k subsets).
- Quota: 950,000 image rows over FineVision's 8 image categories in proportion to sqrt(available rows), then over the subsets of a category the same way, capped at 50,000 rows per subset; 50,000 text-only rows. Whole row groups are drawn at random.
- Deduplication: exact duplicates (same conversation and same image bytes) are removed across the whole set -- 11,377 among the candidates; rows are backfilled from unused passing rows of the same subset, or of the same category where a subset ran out (mostly scienceqa(nona_context)), so the category totals stay at the quota.
| category | rows |
|---|---|
| General VQA | 185,659 |
| Captioning & Knowledge | 184,633 |
| Naive OCR | 135,203 |
| Chart & Table | 113,330 |
| OCR QA | 106,425 |
| Grounding & Counting | 83,126 |
| Mathematics | 74,159 |
| Science | 67,465 |
| Text-only | 50,000 |
| total | 1,000,000 (163 subsets) |
Format
data/category=<category>/train-XXXXX-of-YYYYY.parquet, ~1 GiB shards, 100-row row
groups, rows shuffled within each category.
| column | content |
|---|---|
conversations |
list of {"from": "human"/"gpt", "value"}; "<image>\n" x n_images is prepended to the first human turn (two subsets already carry exactly n_images "<image>" and are kept as they are); a literal <image>/<video> inside the text is escaped to <image >/<video > |
images |
list of Image(decode=False) (bytes, path); the source bytes, never re-encoded |
data_path |
"" |
category, subset |
FineVision category / subset |
source_file, source_row_group, source_row |
where the row lives in FineVision |
relevance_min, visual_dependency_min, image_correspondence_min, formatting_min |
FineVision ratings (null for text-only) |
image_sha256 |
sha256 of each image's bytes |
Verification
Checked against the source with tools/pack/verify_finevision_v2.py
(github.com/williamium3000/Qwenvl2d5): every row re-read from its FineVision row group --
conversation (conversion re-implemented independently), every image's sha256 and the
ratings equal, the selection rules re-derived from the source rows; every image decodes;
no exact duplicate; category totals as above; 306 random rows through the training
preprocessing (median 599 tokens, max 2574).
- Downloads last month
- 1,644