| { |
| "merge": { |
| "base_model": "microsoft/Florence-2-large", |
| "adapter_dir": "/data/anime-captioner/checkpoints/florence2_cumulative_teacher_v2_extension_late_stage_v1/best", |
| "output_dir": "/data/anime-captioner/hf.co/florence2-cumulative-teacher-v2p1", |
| "merged": true, |
| "safe_serialization": true, |
| "no_adapter_needed": true |
| }, |
| "items": [ |
| { |
| "image_id": "dataset/danbooru_images/danbooru_8428374.jpg", |
| "example_file": "examples/01_danbooru_8428374.jpg", |
| "source": "wd14_tags_danbooru", |
| "adult_status": "sensitive", |
| "gap_category": "action", |
| "salience_class": "concept", |
| "wd14_tags": [ |
| "weapon (0.88)", |
| "long hair (0.88)", |
| "sword (0.85)", |
| "animal ears (0.85)", |
| "holding weapon (0.68)", |
| "fox ears (0.67)", |
| "holding (0.67)", |
| "holding sword (0.65)", |
| "katana (0.61)", |
| "skirt (0.59)" |
| ], |
| "wd14_hint_tags": [ |
| { |
| "name": "weapon", |
| "score": 0.8809, |
| "category": 0 |
| }, |
| { |
| "name": "long_hair", |
| "score": 0.875, |
| "category": 0 |
| }, |
| { |
| "name": "sword", |
| "score": 0.8496, |
| "category": 0 |
| }, |
| { |
| "name": "animal_ears", |
| "score": 0.8477, |
| "category": 0 |
| }, |
| { |
| "name": "holding_weapon", |
| "score": 0.6831, |
| "category": 0 |
| }, |
| { |
| "name": "fox_ears", |
| "score": 0.6699, |
| "category": 0 |
| }, |
| { |
| "name": "holding", |
| "score": 0.665, |
| "category": 0 |
| }, |
| { |
| "name": "holding_sword", |
| "score": 0.6519, |
| "category": 0 |
| }, |
| { |
| "name": "katana", |
| "score": 0.6123, |
| "category": 0 |
| }, |
| { |
| "name": "skirt", |
| "score": 0.5942, |
| "category": 0 |
| }, |
| { |
| "name": "1girl", |
| "score": 0.5454, |
| "category": 0 |
| }, |
| { |
| "name": "tail", |
| "score": 0.5376, |
| "category": 0 |
| }, |
| { |
| "name": "fox_tail", |
| "score": 0.4358, |
| "category": 0 |
| }, |
| { |
| "name": "multiple_girls", |
| "score": 0.4224, |
| "category": 0 |
| }, |
| { |
| "name": "2girls", |
| "score": 0.3923, |
| "category": 0 |
| }, |
| { |
| "name": "outdoors", |
| "score": 0.373, |
| "category": 0 |
| }, |
| { |
| "name": "fox_girl", |
| "score": 0.3721, |
| "category": 0 |
| } |
| ], |
| "baseline_short": "An illustration of a woman with a sword in a city.", |
| "baseline_detail": "The image shows an animated person holding a sword in front of a cityscape, with buildings in the background and a watermark on the image.", |
| "current_short": "Dark-haired horned warrior in dark armor wielding a large sword amid blue energy effects in a ruined city.", |
| "current_detail": "A dark-haired figure with horns and a flowing black cloak stands in a dynamic pose, gripping a large sword with both hands, surrounded by swirling blue energy effects and debris in a ruined cityscape with broken buildings and a pink flower-like structure floating in the air.", |
| "baseline_short_words": 11, |
| "baseline_detail_words": 25, |
| "current_short_words": 19, |
| "current_detail_words": 48 |
| }, |
| { |
| "image_id": "dataset/danbooru_images/danbooru_5341516.jpg", |
| "example_file": "examples/02_danbooru_5341516.jpg", |
| "source": "wd14_tags_danbooru", |
| "adult_status": "general", |
| "gap_category": "interactions", |
| "salience_class": "concept", |
| "wd14_tags": [ |
| "mostima (arknights) (0.97)", |
| "multiple girls (0.91)", |
| "2girls (0.91)", |
| "doughnut (0.90)", |
| "food (0.88)", |
| "? (0.87)", |
| "tail (0.86)", |
| "horns (0.86)", |
| "halo (0.85)", |
| "hug from behind (0.83)" |
| ], |
| "wd14_hint_tags": [ |
| { |
| "name": "mostima_(arknights)", |
| "score": 0.9731, |
| "category": 4 |
| }, |
| { |
| "name": "multiple_girls", |
| "score": 0.9136, |
| "category": 0 |
| }, |
| { |
| "name": "2girls", |
| "score": 0.9072, |
| "category": 0 |
| }, |
| { |
| "name": "doughnut", |
| "score": 0.9014, |
| "category": 0 |
| }, |
| { |
| "name": "food", |
| "score": 0.8833, |
| "category": 0 |
| }, |
| { |
| "name": "?", |
| "score": 0.8672, |
| "category": 0 |
| }, |
| { |
| "name": "tail", |
| "score": 0.8647, |
| "category": 0 |
| }, |
| { |
| "name": "horns", |
| "score": 0.8569, |
| "category": 0 |
| }, |
| { |
| "name": "halo", |
| "score": 0.8452, |
| "category": 0 |
| }, |
| { |
| "name": "hug_from_behind", |
| "score": 0.832, |
| "category": 0 |
| }, |
| { |
| "name": "detached_wings", |
| "score": 0.8203, |
| "category": 0 |
| }, |
| { |
| "name": "holding", |
| "score": 0.7744, |
| "category": 0 |
| }, |
| { |
| "name": "blue_hair", |
| "score": 0.769, |
| "category": 0 |
| }, |
| { |
| "name": "black_shorts", |
| "score": 0.7368, |
| "category": 0 |
| }, |
| { |
| "name": "demon_horns", |
| "score": 0.7095, |
| "category": 0 |
| }, |
| { |
| "name": "hug", |
| "score": 0.7061, |
| "category": 0 |
| }, |
| { |
| "name": "holding_food", |
| "score": 0.7017, |
| "category": 0 |
| }, |
| { |
| "name": "hood", |
| "score": 0.7012, |
| "category": 0 |
| }, |
| { |
| "name": "blue_eyes", |
| "score": 0.6968, |
| "category": 0 |
| }, |
| { |
| "name": "long_hair", |
| "score": 0.6934, |
| "category": 0 |
| }, |
| { |
| "name": "demon_tail", |
| "score": 0.6582, |
| "category": 0 |
| }, |
| { |
| "name": "brown_hair", |
| "score": 0.6504, |
| "category": 0 |
| }, |
| { |
| "name": "shorts", |
| "score": 0.6489, |
| "category": 0 |
| }, |
| { |
| "name": "wings", |
| "score": 0.6382, |
| "category": 0 |
| }, |
| { |
| "name": "yuri", |
| "score": 0.6323, |
| "category": 0 |
| }, |
| { |
| "name": "short_hair", |
| "score": 0.5938, |
| "category": 0 |
| }, |
| { |
| "name": "simple_background", |
| "score": 0.5938, |
| "category": 0 |
| }, |
| { |
| "name": "open_mouth", |
| "score": 0.5742, |
| "category": 0 |
| }, |
| { |
| "name": "smile", |
| "score": 0.5698, |
| "category": 0 |
| }, |
| { |
| "name": "hood_down", |
| "score": 0.5347, |
| "category": 0 |
| }, |
| { |
| "name": "sitting", |
| "score": 0.5288, |
| "category": 0 |
| }, |
| { |
| "name": "long_sleeves", |
| "score": 0.4817, |
| "category": 0 |
| }, |
| { |
| "name": "shirt", |
| "score": 0.4702, |
| "category": 0 |
| }, |
| { |
| "name": "green_eyes", |
| "score": 0.467, |
| "category": 0 |
| }, |
| { |
| "name": "colored_skin", |
| "score": 0.4631, |
| "category": 0 |
| }, |
| { |
| "name": "grey_background", |
| "score": 0.4277, |
| "category": 0 |
| }, |
| { |
| "name": "alternate_costume", |
| "score": 0.395, |
| "category": 0 |
| }, |
| { |
| "name": "energy_wings", |
| "score": 0.3916, |
| "category": 0 |
| }, |
| { |
| "name": "hoodie", |
| "score": 0.3596, |
| "category": 0 |
| } |
| ], |
| "baseline_short": "A couple of anime girls sitting next to each other.", |
| "baseline_detail": "The image shows two anime girls sitting next to each other on a couch, with one of them holding a donut in her hand. The image is animated, giving it a lively and vibrant feel.", |
| "current_short": "Blue-haired cat-eared girl holds a donut beside a red-haired horned girl in a hoodie.", |
| "current_detail": "A blue-haired girl with cat ears and a halo holds a donut in her right hand while resting her left hand on the lap of a pink-haired horned girl in a black hoodie, their arms wrapped around each other in a close embrace.", |
| "baseline_short_words": 10, |
| "baseline_detail_words": 35, |
| "current_short_words": 17, |
| "current_detail_words": 45 |
| }, |
| { |
| "image_id": "dataset/danbooru_images/danbooru_5294973.jpg", |
| "example_file": "examples/03_danbooru_5294973.jpg", |
| "source": "wd14_tags_danbooru", |
| "adult_status": "general", |
| "gap_category": "body_diversity", |
| "salience_class": "concept", |
| "wd14_tags": [ |
| "guts (berserk) (0.97)", |
| "armor (0.92)", |
| "multiple boys (0.92)", |
| "2boys (0.92)", |
| "weapon (0.88)", |
| "long hair (0.83)", |
| "sword (0.78)", |
| "blue skin (0.74)", |
| "colored skin (0.73)", |
| "wavy hair (0.70)" |
| ], |
| "wd14_hint_tags": [ |
| { |
| "name": "guts_(berserk)", |
| "score": 0.9653, |
| "category": 4 |
| }, |
| { |
| "name": "armor", |
| "score": 0.9199, |
| "category": 0 |
| }, |
| { |
| "name": "multiple_boys", |
| "score": 0.9194, |
| "category": 0 |
| }, |
| { |
| "name": "2boys", |
| "score": 0.916, |
| "category": 0 |
| }, |
| { |
| "name": "weapon", |
| "score": 0.8843, |
| "category": 0 |
| }, |
| { |
| "name": "long_hair", |
| "score": 0.8301, |
| "category": 0 |
| }, |
| { |
| "name": "sword", |
| "score": 0.7764, |
| "category": 0 |
| }, |
| { |
| "name": "blue_skin", |
| "score": 0.7417, |
| "category": 0 |
| }, |
| { |
| "name": "colored_skin", |
| "score": 0.7319, |
| "category": 0 |
| }, |
| { |
| "name": "wavy_hair", |
| "score": 0.7021, |
| "category": 0 |
| }, |
| { |
| "name": "white_hair", |
| "score": 0.7007, |
| "category": 0 |
| }, |
| { |
| "name": "1girl", |
| "score": 0.6919, |
| "category": 0 |
| }, |
| { |
| "name": "short_hair", |
| "score": 0.6738, |
| "category": 0 |
| }, |
| { |
| "name": "white_background", |
| "score": 0.5962, |
| "category": 0 |
| }, |
| { |
| "name": "black_hair", |
| "score": 0.5825, |
| "category": 0 |
| }, |
| { |
| "name": "cape", |
| "score": 0.5601, |
| "category": 0 |
| }, |
| { |
| "name": "closed_mouth", |
| "score": 0.5293, |
| "category": 0 |
| }, |
| { |
| "name": "holding_weapon", |
| "score": 0.4985, |
| "category": 0 |
| }, |
| { |
| "name": "holding", |
| "score": 0.4763, |
| "category": 0 |
| }, |
| { |
| "name": "simple_background", |
| "score": 0.4524, |
| "category": 0 |
| }, |
| { |
| "name": "shoulder_armor", |
| "score": 0.4397, |
| "category": 0 |
| }, |
| { |
| "name": "androgynous", |
| "score": 0.3752, |
| "category": 0 |
| }, |
| { |
| "name": "breastplate", |
| "score": 0.3723, |
| "category": 0 |
| }, |
| { |
| "name": "holding_sword", |
| "score": 0.3708, |
| "category": 0 |
| } |
| ], |
| "baseline_short": "A group of anime characters standing next to each other.", |
| "baseline_detail": "The image shows a group of four people standing next to each other in front of a white background. The person in the front is wearing a white dress and is holding a sword, while the person to the left is wearing an unknown outfit. The image is animated, giving it a dynamic feel.", |
| "current_short": "White-haired woman in silver armor flanked by two dark-haired men in red robes.", |
| "current_detail": "A woman with long white curly hair and blue eyes stands in the foreground wearing a white and silver armor with a glowing blue orb at her chest. To her left is a young man with short black hair and a red jacket, and to her right is a man with spiky dark hair wearing a red cape and holding a sword.", |
| "baseline_short_words": 10, |
| "baseline_detail_words": 54, |
| "current_short_words": 15, |
| "current_detail_words": 62 |
| }, |
| { |
| "image_id": "dataset/danbooru_images/danbooru_7019581.jpg", |
| "example_file": "examples/04_danbooru_7019581.jpg", |
| "source": "wd14_tags_danbooru", |
| "adult_status": "general", |
| "gap_category": "environments", |
| "salience_class": "concept", |
| "wd14_tags": [ |
| "1girl (0.94)", |
| "scarf (0.91)", |
| "solo (0.89)", |
| "lying (0.85)", |
| "animal ears (0.85)", |
| "halo (0.85)", |
| "one eye closed (0.82)", |
| "on back (0.82)", |
| "winter clothes (0.73)", |
| "smile (0.68)" |
| ], |
| "wd14_hint_tags": [ |
| { |
| "name": "1girl", |
| "score": 0.9448, |
| "category": 0 |
| }, |
| { |
| "name": "scarf", |
| "score": 0.9136, |
| "category": 0 |
| }, |
| { |
| "name": "solo", |
| "score": 0.8877, |
| "category": 0 |
| }, |
| { |
| "name": "lying", |
| "score": 0.8521, |
| "category": 0 |
| }, |
| { |
| "name": "animal_ears", |
| "score": 0.8511, |
| "category": 0 |
| }, |
| { |
| "name": "halo", |
| "score": 0.8491, |
| "category": 0 |
| }, |
| { |
| "name": "one_eye_closed", |
| "score": 0.8198, |
| "category": 0 |
| }, |
| { |
| "name": "on_back", |
| "score": 0.8193, |
| "category": 0 |
| }, |
| { |
| "name": "winter_clothes", |
| "score": 0.7349, |
| "category": 0 |
| }, |
| { |
| "name": "smile", |
| "score": 0.6826, |
| "category": 0 |
| }, |
| { |
| "name": "coat", |
| "score": 0.6777, |
| "category": 0 |
| }, |
| { |
| "name": "fox_ears", |
| "score": 0.6772, |
| "category": 0 |
| }, |
| { |
| "name": "tail", |
| "score": 0.6709, |
| "category": 0 |
| }, |
| { |
| "name": "snow", |
| "score": 0.6445, |
| "category": 0 |
| }, |
| { |
| "name": "snowball", |
| "score": 0.6445, |
| "category": 0 |
| }, |
| { |
| "name": "black_hair", |
| "score": 0.6211, |
| "category": 0 |
| }, |
| { |
| "name": "boots", |
| "score": 0.6162, |
| "category": 0 |
| }, |
| { |
| "name": "blue_eyes", |
| "score": 0.6099, |
| "category": 0 |
| }, |
| { |
| "name": "purple_scarf", |
| "score": 0.5894, |
| "category": 0 |
| }, |
| { |
| "name": "fox_tail", |
| "score": 0.5522, |
| "category": 0 |
| }, |
| { |
| "name": "looking_at_viewer", |
| "score": 0.5264, |
| "category": 0 |
| }, |
| { |
| "name": "long_sleeves", |
| "score": 0.5073, |
| "category": 0 |
| }, |
| { |
| "name": "open_mouth", |
| "score": 0.5005, |
| "category": 0 |
| }, |
| { |
| "name": "fox_girl", |
| "score": 0.4875, |
| "category": 0 |
| }, |
| { |
| "name": "fang", |
| "score": 0.4753, |
| "category": 0 |
| }, |
| { |
| "name": "blush", |
| "score": 0.4697, |
| "category": 0 |
| }, |
| { |
| "name": "mittens", |
| "score": 0.4661, |
| "category": 0 |
| }, |
| { |
| "name": "white_coat", |
| "score": 0.4622, |
| "category": 0 |
| }, |
| { |
| "name": "animal_ear_fluff", |
| "score": 0.4458, |
| "category": 0 |
| }, |
| { |
| "name": "bangs", |
| "score": 0.4238, |
| "category": 0 |
| } |
| ], |
| "baseline_short": "A girl in a white coat laying on a pillow.", |
| "baseline_detail": "The image shows a girl in a white coat laying on top of a bed next to a cat, with a pillow beneath her head. She appears to be sleeping peacefully, with her eyes closed and her arms tucked in close to her body. The background is a soft, muted color, giving the image a dreamy, ethereal feel.", |
| "current_short": "Cat-eared girl in a white coat lies on a bed with a fluffy tail.", |
| "current_detail": "A cat-eared girl with brown and white hair lies on her back across white sheets, eyes closed and mouth open, wearing a white puffy coat, pink scarf, black gloves, and brown boots. A large fluffy tail rests between her legs, and a pink circular emblem floats above her head.", |
| "baseline_short_words": 10, |
| "baseline_detail_words": 58, |
| "current_short_words": 15, |
| "current_detail_words": 50 |
| }, |
| { |
| "image_id": "dataset/danbooru_images/danbooru_8654121.jpg", |
| "example_file": "examples/05_danbooru_8654121.jpg", |
| "source": "wd14_tags_danbooru", |
| "adult_status": "sensitive", |
| "gap_category": "everyday", |
| "salience_class": "concept", |
| "wd14_tags": [ |
| "1girl (0.99)", |
| "japanese clothes (0.96)", |
| "kimono (0.95)", |
| "umbrella (0.94)", |
| "solo (0.89)", |
| "animal ears (0.88)", |
| "looking back (0.80)", |
| "sword (0.79)", |
| "weapon (0.79)", |
| "from behind (0.78)" |
| ], |
| "wd14_hint_tags": [ |
| { |
| "name": "1girl", |
| "score": 0.9888, |
| "category": 0 |
| }, |
| { |
| "name": "japanese_clothes", |
| "score": 0.9595, |
| "category": 0 |
| }, |
| { |
| "name": "kimono", |
| "score": 0.9487, |
| "category": 0 |
| }, |
| { |
| "name": "umbrella", |
| "score": 0.9351, |
| "category": 0 |
| }, |
| { |
| "name": "solo", |
| "score": 0.8877, |
| "category": 0 |
| }, |
| { |
| "name": "animal_ears", |
| "score": 0.8799, |
| "category": 0 |
| }, |
| { |
| "name": "looking_back", |
| "score": 0.8018, |
| "category": 0 |
| }, |
| { |
| "name": "sword", |
| "score": 0.7891, |
| "category": 0 |
| }, |
| { |
| "name": "weapon", |
| "score": 0.7871, |
| "category": 0 |
| }, |
| { |
| "name": "from_behind", |
| "score": 0.7837, |
| "category": 0 |
| }, |
| { |
| "name": "sitting", |
| "score": 0.7603, |
| "category": 0 |
| }, |
| { |
| "name": "blurry_background", |
| "score": 0.7598, |
| "category": 0 |
| }, |
| { |
| "name": "looking_at_viewer", |
| "score": 0.7505, |
| "category": 0 |
| }, |
| { |
| "name": "blurry", |
| "score": 0.7437, |
| "category": 0 |
| }, |
| { |
| "name": "sash", |
| "score": 0.7095, |
| "category": 0 |
| }, |
| { |
| "name": "fox_ears", |
| "score": 0.7046, |
| "category": 0 |
| }, |
| { |
| "name": "katana", |
| "score": 0.6973, |
| "category": 0 |
| }, |
| { |
| "name": "braid", |
| "score": 0.6655, |
| "category": 0 |
| }, |
| { |
| "name": "bangs", |
| "score": 0.6528, |
| "category": 0 |
| }, |
| { |
| "name": "oil-paper_umbrella", |
| "score": 0.6401, |
| "category": 0 |
| }, |
| { |
| "name": "obi", |
| "score": 0.6367, |
| "category": 0 |
| }, |
| { |
| "name": "virtual_youtuber", |
| "score": 0.6265, |
| "category": 0 |
| }, |
| { |
| "name": "purple_eyes", |
| "score": 0.6216, |
| "category": 0 |
| }, |
| { |
| "name": "holding", |
| "score": 0.5474, |
| "category": 0 |
| }, |
| { |
| "name": "hair_bun", |
| "score": 0.5386, |
| "category": 0 |
| }, |
| { |
| "name": "brown_hair", |
| "score": 0.5239, |
| "category": 0 |
| }, |
| { |
| "name": "single_hair_bun", |
| "score": 0.5195, |
| "category": 0 |
| }, |
| { |
| "name": "food", |
| "score": 0.5103, |
| "category": 0 |
| }, |
| { |
| "name": "sheath", |
| "score": 0.5078, |
| "category": 0 |
| }, |
| { |
| "name": "outdoors", |
| "score": 0.501, |
| "category": 0 |
| }, |
| { |
| "name": "blue_eyes", |
| "score": 0.4854, |
| "category": 0 |
| }, |
| { |
| "name": "blunt_bangs", |
| "score": 0.4744, |
| "category": 0 |
| }, |
| { |
| "name": "depth_of_field", |
| "score": 0.4468, |
| "category": 0 |
| }, |
| { |
| "name": "floral_print", |
| "score": 0.4407, |
| "category": 0 |
| }, |
| { |
| "name": "print_kimono", |
| "score": 0.4089, |
| "category": 0 |
| }, |
| { |
| "name": "long_hair", |
| "score": 0.4009, |
| "category": 0 |
| }, |
| { |
| "name": "yukata", |
| "score": 0.3801, |
| "category": 0 |
| }, |
| { |
| "name": "sheathed", |
| "score": 0.373, |
| "category": 0 |
| }, |
| { |
| "name": "yellow_kimono", |
| "score": 0.356, |
| "category": 0 |
| } |
| ], |
| "baseline_short": "A woman in a kimono holding an umbrella and a cat.", |
| "baseline_detail": "The image shows a woman in a kimono sitting on a bench, holding an umbrella in one hand and a stick in the other. She is surrounded by trees and a shed in the background, creating a peaceful atmosphere. The image is animated, giving it a dynamic feel.", |
| "current_short": "Cat-eared girl in a blue floral kimono holding a red umbrella.", |
| "current_detail": "A young woman with black hair in a high bun and fox ears wears a teal kimono patterned with white and gold floral designs while seated on a wooden surface. She holds a small red object in one hand and rests the other on her lap, with a red umbrella visible behind her and a sword resting on the ground in front of her.", |
| "baseline_short_words": 11, |
| "baseline_detail_words": 48, |
| "current_short_words": 12, |
| "current_detail_words": 64 |
| }, |
| { |
| "image_id": "dataset/danbooru_images/danbooru_5221484.jpg", |
| "example_file": "examples/06_danbooru_5221484.jpg", |
| "source": "wd14_tags_danbooru", |
| "adult_status": "general", |
| "gap_category": "body_diversity", |
| "salience_class": "concept", |
| "wd14_tags": [ |
| "hina (blue archive) (0.99)", |
| "1girl (0.97)", |
| "hina (swimsuit) (blue archive) (0.96)", |
| "whistle (0.95)", |
| "innertube (0.93)", |
| "horns (0.92)", |
| "purple eyes (0.85)", |
| "ahoge (0.85)", |
| "long hair (0.84)", |
| "hair ornament (0.82)" |
| ], |
| "wd14_hint_tags": [ |
| { |
| "name": "hina_(blue_archive)", |
| "score": 0.9946, |
| "category": 4 |
| }, |
| { |
| "name": "1girl", |
| "score": 0.9712, |
| "category": 0 |
| }, |
| { |
| "name": "hina_(swimsuit)_(blue_archive)", |
| "score": 0.9575, |
| "category": 4 |
| }, |
| { |
| "name": "whistle", |
| "score": 0.9482, |
| "category": 0 |
| }, |
| { |
| "name": "innertube", |
| "score": 0.9316, |
| "category": 0 |
| }, |
| { |
| "name": "horns", |
| "score": 0.9243, |
| "category": 0 |
| }, |
| { |
| "name": "purple_eyes", |
| "score": 0.855, |
| "category": 0 |
| }, |
| { |
| "name": "ahoge", |
| "score": 0.8535, |
| "category": 0 |
| }, |
| { |
| "name": "long_hair", |
| "score": 0.8403, |
| "category": 0 |
| }, |
| { |
| "name": "hair_ornament", |
| "score": 0.8164, |
| "category": 0 |
| }, |
| { |
| "name": "solo", |
| "score": 0.8086, |
| "category": 0 |
| }, |
| { |
| "name": "hairclip", |
| "score": 0.7891, |
| "category": 0 |
| }, |
| { |
| "name": "halo", |
| "score": 0.7866, |
| "category": 0 |
| }, |
| { |
| "name": "bangs", |
| "score": 0.6855, |
| "category": 0 |
| }, |
| { |
| "name": "demon_horns", |
| "score": 0.6587, |
| "category": 0 |
| }, |
| { |
| "name": "english_text", |
| "score": 0.6582, |
| "category": 0 |
| }, |
| { |
| "name": "chibi", |
| "score": 0.6172, |
| "category": 0 |
| }, |
| { |
| "name": "arms_up", |
| "score": 0.5898, |
| "category": 0 |
| }, |
| { |
| "name": "white_background", |
| "score": 0.5679, |
| "category": 0 |
| }, |
| { |
| "name": "speech_bubble", |
| "score": 0.5405, |
| "category": 0 |
| }, |
| { |
| "name": "swimsuit", |
| "score": 0.5312, |
| "category": 0 |
| }, |
| { |
| "name": "simple_background", |
| "score": 0.5156, |
| "category": 0 |
| }, |
| { |
| "name": "hair_between_eyes", |
| "score": 0.4978, |
| "category": 0 |
| }, |
| { |
| "name": "multiple_horns", |
| "score": 0.4863, |
| "category": 0 |
| }, |
| { |
| "name": "whistle_around_neck", |
| "score": 0.4219, |
| "category": 0 |
| }, |
| { |
| "name": "white_hair", |
| "score": 0.4192, |
| "category": 0 |
| }, |
| { |
| "name": "parted_bangs", |
| "score": 0.4004, |
| "category": 0 |
| }, |
| { |
| "name": "colored_skin", |
| "score": 0.3672, |
| "category": 0 |
| } |
| ], |
| "baseline_short": "A cartoon of a girl floating in a pool with a duck.", |
| "baseline_detail": "The image shows a cartoon of a girl floating in the water with a crown on her head, wearing a life jacket and holding an object in her hand. The background is white and there is some text at the bottom of the image.", |
| "current_short": "White-haired horned girl in an inflatable ring with a white duck floating in blue water.", |
| "current_detail": "A white-haired girl with spiky hair and a red bandage floats in blue water, her face flushed with eyes closed and mouth open, while a small white creature with a pink bow floats beside her and a purple crown floats above her head.", |
| "baseline_short_words": 12, |
| "baseline_detail_words": 44, |
| "current_short_words": 16, |
| "current_detail_words": 44 |
| }, |
| { |
| "image_id": "dataset/danbooru_images/danbooru_8643115.jpg", |
| "example_file": "examples/07_danbooru_8643115.jpg", |
| "source": "wd14_tags_danbooru", |
| "adult_status": "general", |
| "gap_category": "fantasy", |
| "salience_class": "concept", |
| "wd14_tags": [ |
| "multiple girls (0.98)", |
| "2girls (0.96)", |
| "kimono (0.96)", |
| "japanese clothes (0.95)", |
| "wings (0.93)", |
| "seiza (0.91)", |
| "hina (blue archive) (0.91)", |
| "sash (0.89)", |
| "halo (0.87)", |
| "sitting (0.85)" |
| ], |
| "wd14_hint_tags": [ |
| { |
| "name": "multiple_girls", |
| "score": 0.9766, |
| "category": 0 |
| }, |
| { |
| "name": "2girls", |
| "score": 0.9556, |
| "category": 0 |
| }, |
| { |
| "name": "kimono", |
| "score": 0.9551, |
| "category": 0 |
| }, |
| { |
| "name": "japanese_clothes", |
| "score": 0.9492, |
| "category": 0 |
| }, |
| { |
| "name": "wings", |
| "score": 0.9346, |
| "category": 0 |
| }, |
| { |
| "name": "seiza", |
| "score": 0.9102, |
| "category": 0 |
| }, |
| { |
| "name": "hina_(blue_archive)", |
| "score": 0.9072, |
| "category": 4 |
| }, |
| { |
| "name": "sash", |
| "score": 0.8862, |
| "category": 0 |
| }, |
| { |
| "name": "halo", |
| "score": 0.8677, |
| "category": 0 |
| }, |
| { |
| "name": "sitting", |
| "score": 0.8511, |
| "category": 0 |
| }, |
| { |
| "name": "obi", |
| "score": 0.8286, |
| "category": 0 |
| }, |
| { |
| "name": "white_hair", |
| "score": 0.8271, |
| "category": 0 |
| }, |
| { |
| "name": "purple_eyes", |
| "score": 0.8115, |
| "category": 0 |
| }, |
| { |
| "name": "smile", |
| "score": 0.7847, |
| "category": 0 |
| }, |
| { |
| "name": "hair_ornament", |
| "score": 0.7603, |
| "category": 0 |
| }, |
| { |
| "name": "ahoge", |
| "score": 0.7559, |
| "category": 0 |
| }, |
| { |
| "name": "looking_at_viewer", |
| "score": 0.7397, |
| "category": 0 |
| }, |
| { |
| "name": "hair_flower", |
| "score": 0.7305, |
| "category": 0 |
| }, |
| { |
| "name": "long_hair", |
| "score": 0.7188, |
| "category": 0 |
| }, |
| { |
| "name": "flower", |
| "score": 0.7134, |
| "category": 0 |
| }, |
| { |
| "name": "horns", |
| "score": 0.7119, |
| "category": 0 |
| }, |
| { |
| "name": "demon_wings", |
| "score": 0.6782, |
| "category": 0 |
| }, |
| { |
| "name": "wide_sleeves", |
| "score": 0.6455, |
| "category": 0 |
| }, |
| { |
| "name": "heterochromia", |
| "score": 0.644, |
| "category": 0 |
| }, |
| { |
| "name": "bangs", |
| "score": 0.5815, |
| "category": 0 |
| }, |
| { |
| "name": "multiple_horns", |
| "score": 0.5503, |
| "category": 0 |
| }, |
| { |
| "name": "demon_horns", |
| "score": 0.5381, |
| "category": 0 |
| }, |
| { |
| "name": "blue_eyes", |
| "score": 0.5244, |
| "category": 0 |
| }, |
| { |
| "name": "gloves", |
| "score": 0.519, |
| "category": 0 |
| }, |
| { |
| "name": "floral_print", |
| "score": 0.5034, |
| "category": 0 |
| }, |
| { |
| "name": "long_sleeves", |
| "score": 0.4763, |
| "category": 0 |
| }, |
| { |
| "name": "alternate_costume", |
| "score": 0.4624, |
| "category": 0 |
| }, |
| { |
| "name": "black_gloves", |
| "score": 0.458, |
| "category": 0 |
| }, |
| { |
| "name": "sidelocks", |
| "score": 0.4465, |
| "category": 0 |
| }, |
| { |
| "name": "parted_bangs", |
| "score": 0.4465, |
| "category": 0 |
| }, |
| { |
| "name": "purple_hair", |
| "score": 0.4062, |
| "category": 0 |
| }, |
| { |
| "name": "hair_bun", |
| "score": 0.3994, |
| "category": 0 |
| }, |
| { |
| "name": "open_mouth", |
| "score": 0.3701, |
| "category": 0 |
| }, |
| { |
| "name": "sliding_doors", |
| "score": 0.3691, |
| "category": 0 |
| } |
| ], |
| "baseline_short": "A couple of girls sitting on top of a tatami mat.", |
| "baseline_detail": "The image shows three anime girls in kimono sitting on the floor in front of a window, with a houseplant on the left side and text on the image. Through the window, we can see trees and snow outside.", |
| "current_short": "Pink-haired girl in a white kimono and white-haired dragon-eared girl in purple kimonos sit on a tatami mat.", |
| "current_detail": "A pink-haired girl in a white kimono with a pink bow and floral headpiece sits on the left, holding a pink bowl, while a white-haired dragon-like girl with purple eyes kneels on the right in a purple kimonos with black wings and a white headpiece, both positioned on yellow tatami mats in front of large windows with snow-covered trees visible outside.", |
| "baseline_short_words": 11, |
| "baseline_detail_words": 39, |
| "current_short_words": 21, |
| "current_detail_words": 65 |
| }, |
| { |
| "image_id": "dataset/danbooru_images/danbooru_6621403.jpg", |
| "example_file": "examples/08_danbooru_6621403.jpg", |
| "source": "wd14_tags_danbooru", |
| "adult_status": "general", |
| "gap_category": "environments", |
| "salience_class": "concept", |
| "wd14_tags": [ |
| "1girl (0.99)", |
| "solo (0.97)", |
| "animal ears (0.95)", |
| "sword (0.95)", |
| "weapon (0.94)", |
| "holding sword (0.93)", |
| "holding (0.93)", |
| "holding weapon (0.92)", |
| "long hair (0.86)", |
| "bear ears (0.86)" |
| ], |
| "wd14_hint_tags": [ |
| { |
| "name": "1girl", |
| "score": 0.9917, |
| "category": 0 |
| }, |
| { |
| "name": "solo", |
| "score": 0.9702, |
| "category": 0 |
| }, |
| { |
| "name": "animal_ears", |
| "score": 0.9497, |
| "category": 0 |
| }, |
| { |
| "name": "sword", |
| "score": 0.9487, |
| "category": 0 |
| }, |
| { |
| "name": "weapon", |
| "score": 0.9365, |
| "category": 0 |
| }, |
| { |
| "name": "holding_sword", |
| "score": 0.9321, |
| "category": 0 |
| }, |
| { |
| "name": "holding", |
| "score": 0.9282, |
| "category": 0 |
| }, |
| { |
| "name": "holding_weapon", |
| "score": 0.9229, |
| "category": 0 |
| }, |
| { |
| "name": "long_hair", |
| "score": 0.8574, |
| "category": 0 |
| }, |
| { |
| "name": "bear_ears", |
| "score": 0.8569, |
| "category": 0 |
| }, |
| { |
| "name": "sitting", |
| "score": 0.8062, |
| "category": 0 |
| }, |
| { |
| "name": "black_hairband", |
| "score": 0.7856, |
| "category": 0 |
| }, |
| { |
| "name": "black_footwear", |
| "score": 0.7783, |
| "category": 0 |
| }, |
| { |
| "name": "hairband", |
| "score": 0.772, |
| "category": 0 |
| }, |
| { |
| "name": "long_sleeves", |
| "score": 0.7539, |
| "category": 0 |
| }, |
| { |
| "name": "outdoors", |
| "score": 0.7495, |
| "category": 0 |
| }, |
| { |
| "name": "looking_at_viewer", |
| "score": 0.7295, |
| "category": 0 |
| }, |
| { |
| "name": "planted_sword", |
| "score": 0.7275, |
| "category": 0 |
| }, |
| { |
| "name": "wariza", |
| "score": 0.6968, |
| "category": 0 |
| }, |
| { |
| "name": "sky", |
| "score": 0.6836, |
| "category": 0 |
| }, |
| { |
| "name": "pantyhose", |
| "score": 0.6406, |
| "category": 0 |
| }, |
| { |
| "name": "star_(sky)", |
| "score": 0.6333, |
| "category": 0 |
| }, |
| { |
| "name": "bangs", |
| "score": 0.6143, |
| "category": 0 |
| }, |
| { |
| "name": "purple_hair", |
| "score": 0.6035, |
| "category": 0 |
| }, |
| { |
| "name": "starry_sky", |
| "score": 0.5947, |
| "category": 0 |
| }, |
| { |
| "name": "skirt", |
| "score": 0.5825, |
| "category": 0 |
| }, |
| { |
| "name": "planted", |
| "score": 0.5771, |
| "category": 0 |
| }, |
| { |
| "name": "flower", |
| "score": 0.5723, |
| "category": 0 |
| }, |
| { |
| "name": "belt", |
| "score": 0.5625, |
| "category": 0 |
| }, |
| { |
| "name": "black_pantyhose", |
| "score": 0.543, |
| "category": 0 |
| }, |
| { |
| "name": "closed_mouth", |
| "score": 0.498, |
| "category": 0 |
| }, |
| { |
| "name": "night", |
| "score": 0.4797, |
| "category": 0 |
| }, |
| { |
| "name": "wide_sleeves", |
| "score": 0.4734, |
| "category": 0 |
| }, |
| { |
| "name": "black_dress", |
| "score": 0.4304, |
| "category": 0 |
| }, |
| { |
| "name": "full_body", |
| "score": 0.4224, |
| "category": 0 |
| }, |
| { |
| "name": "thigh_strap", |
| "score": 0.4189, |
| "category": 0 |
| }, |
| { |
| "name": "tail", |
| "score": 0.4158, |
| "category": 0 |
| }, |
| { |
| "name": "twintails", |
| "score": 0.3594, |
| "category": 0 |
| }, |
| { |
| "name": "night_sky", |
| "score": 0.3569, |
| "category": 0 |
| } |
| ], |
| "baseline_short": "A girl in a black dress holding a sword in a field.", |
| "baseline_detail": "The image shows a girl in a black dress holding a sword in a field surrounded by plants with flowers, trees, hills, and a starry sky.", |
| "current_short": "Purple-haired girl with bear ears holds a sword in a field at night.", |
| "current_detail": "A young woman with long purple hair in twin pigtails tied with purple ribbons and teddy bear ears sits on a grassy field at night, wearing a black kimono-style outfit with purple trim and a plaid skirt, holding a tall silver sword in her right hand with her left arm raised.", |
| "baseline_short_words": 12, |
| "baseline_detail_words": 26, |
| "current_short_words": 14, |
| "current_detail_words": 52 |
| } |
| ] |
| } |
|
|
