{"success":true,"data":{"author":"Thames, Karpur, Norris, Xia, Panait, Weyand and Sim (Google Research)","bucketAccessType":"public","bucketListable":true,"bucketName":"nutrition5k_dataset","bucketPrefix":"nutrition5k_dataset","category":"Health & Biomedical","cloudProvider":"gcp","createdAt":"2026-11-19T15:09:40.229Z","description":"Visual and nutritional data for about 5,000 real cafeteria plates scanned in Google cafeterias in California, released with Thames et al., 'Nutrition5k: Towards Automatic Nutritional Understanding of Generic Food' (CVPR 2021). Each of 5,006 dishes was captured on a custom rig with four rotating side-angle videos from Raspberry Pi cameras at 30 and 60 degree angles, plus an overhead RealSense RGB-D image when available (raw 16-bit depth at 10,000 units per meter, and colorized depth). Dishes were built incrementally, one ingredient at a time, and weighed, so labels give a fine-grained ingredient list, per-ingredient mass, total mass and calories, and fat, protein and carbohydrate masses, with per-gram nutrition taken from the USDA Food and Nutrient Database. Listed 2026-11-19: nutrition5k_dataset/ holds imagery/ (side_angles videos, realsense_overhead), metadata/ (dish_metadata_cafe1.csv, dish_metadata_cafe2.csv, ingredient_metadata.csv), dish_ids/ (official train and test splits) and scripts/ (frame extraction and evaluation of calorie and macronutrient regression). A 181 GB tarball of everything sits at the bucket root. Released under CC BY 4.0; cite Thames et al. 2021.","displayTitle":"Nutrition5k: RGB-D Food Imagery with Per-Ingredient Mass, Calories and Macronutrients (Google Research)","fuseMountBrowsable":true,"id":"a11e9e31-39a8-51a6-ac87-ad2a993da22b","license":"CC-BY-4.0","licenseEvidenceUrl":"https://github.com/google-research-datasets/Nutrition5k","objectCountEstimate":4,"probeStatus":"ok","referenceUrl":"https://arxiv.org/abs/2103.03375","region":"us","schemaCard":{"accessHints":[{"language":"python-gcsfs","snippet":"import gcsfs\nfs = gcsfs.GCSFileSystem(token='anon')\nfs.ls('nutrition5k_dataset/nutrition5k_dataset/', detail=False)[:10]"},{"language":"python-storage","snippet":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nlist(client.bucket('nutrition5k_dataset').list_blobs(prefix='nutrition5k_dataset/', max_results=10))"}],"caveats":["Files are binary — use a domain-specific parser rather than text I/O."],"format":"binary","generatedAt":"2026-11-19T15:22:19.747Z","layout":{"example":"nutrition5k_dataset/dish_ids/splits/depth_test_ids.txt","pattern":"{prefix}/{seg1}/{seg2}/{filename}"},"mountPath":"/mnt/gcs-archive/data","sampleHeaders":[{"bytes":"dish_1565035746\ndish_1558639818\ndish_1558549605\ndish_1561663580\ndish_1565898402\ndish_1566328724\ndish_1566838351\ndish_1567107839\ndish_1558639787\ndish_1562788601\ndish_1560456326\ndish_1564427430\ndish_1565896662\ndish_1566414342\ndish_1562617703\ndish_1565020926\ndish_1564689168\ndish_1560456348\ndish_1558641287\ndish_1558373008\ndish_1561577947\ndish_1558639440\ndish_1565894889\ndish_1561147944\ndish_1557937079\ndish_1558641320\ndish_1567007848\ndish_1565810969\ndish_1562168800\ndish_1564687046\ndish_1563221054\ndish_1565898361\n","encoding":"utf8","key":"nutrition5k_dataset/dish_ids/splits/depth_test_ids.txt","truncatedAt":512},{"bytes":"dish_1558641037\ndish_1562958721\ndish_1565799617\ndish_1561491210\ndish_1561661806\ndish_1562873264\ndish_1558373449\ndish_1561480414\ndish_1563465805\ndish_1564589308\ndish_1560442226\ndish_1566333255\ndish_1561134102\ndish_1562602918\ndish_1563898786\ndish_1562615103\ndish_1565711115\ndish_1558031526\ndish_1563475788\ndish_1558725085\ndish_1562168365\ndish_1562945779\ndish_1565194906\ndish_1565808943\ndish_1565023128\ndish_1568044496\ndish_1558457468\ndish_1558114484\ndish_1562704564\ndish_1560368944\ndish_1562946006\ndish_1563218959\n","encoding":"utf8","key":"nutrition5k_dataset/dish_ids/splits/depth_train_ids.txt","truncatedAt":512}],"sampleObjects":["nutrition5k_dataset/dish_ids/splits/depth_test_ids.txt","nutrition5k_dataset/dish_ids/splits/depth_train_ids.txt","nutrition5k_dataset/dish_ids/splits/rgb_test_ids.txt","nutrition5k_dataset/dish_ids/splits/rgb_train_ids.txt"],"sidecars":[{"bytes":"See https://github.com/google-research-datasets/Nutrition5k for details.\n","encoding":"utf8","key":"nutrition5k_dataset/README","name":"README","size":73,"truncated":false},{"bytes":"./dish_ids/README\n\nAll models in the Nutrition5k paper were only trained on data from Cafe 1.\n- Cafe 1 data consists of all dish ids from dish_1556572657 to dish_1568666357\n- Cafe 2 data consists of all dish ids from dish_1571931457 to dish_1576259776\n\nThe ../splits/ directory lists the splits used in our experiments. Note that\nbecause depth information is only available for a subset of the dishes, the\ndepth_*_split.txt files are subsets of the rgb_*_splits.txt files.\n","encoding":"utf8","key":"nutrition5k_dataset/dish_ids/README","name":"README","size":473,"truncated":false}],"sourceObjectKey":"nutrition5k_dataset/dish_ids/splits/depth_test_ids.txt","starterCell":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nbucket = client.bucket('nutrition5k_dataset')\nblob = bucket.blob('nutrition5k_dataset/dish_ids/splits/depth_test_ids.txt')\nwith blob.open(\"rb\") as fh:\n    print(fh.read(64))","starterSamplePath":"dish_ids/splits/depth_test_ids.txt","totalObjects":"4 top-level prefixes"},"summary":"Visual and nutritional data for about 5,000 real cafeteria plates scanned in Google cafeterias in California, released with Thames et al., 'Nutrition5k: Towards Automatic…","title":"Nutrition5k: RGB-D Food Imagery with Per-Ingredient Mass, Calories and Macronutrients (Google Research)","updatedAt":"2026-11-19T19:01:30.588Z","visibility":"public","websiteUrl":"https://github.com/google-research-datasets/Nutrition5k"}}