{"success":true,"data":{"author":"Elizabeth Campolongo, Jianyang Gu, Net Zhang (Imageomics Institute)","bucketAccessType":"public","bucketName":"imageomics/IDLE-OO-Camera-Traps","bucketPrefix":"","category":"Life Sciences","cloudProvider":"huggingface","createdAt":"2026-11-21T20:45:58.661Z","description":"A species classification benchmark of 2,586 camera trap images drawn from five datasets in the Labeled Information Library of Alexandria: Biology and Conservation (LILA BC), each balanced to the same number of images per species, built by the Imageomics Institute to evaluate the BioCLIP 2 biology foundation model in zero-shot, one-shot and five-shot settings. The five sets are Desert Lion Conservation Camera Traps from Namibia (352 images, 32 species), ENA24-detection from eastern North America (1,120 images, 20 species), Island Conservation Camera Traps (310 images, 33 common names), Ohio Small Animals (468 images) and Orinoquia Camera Traps from Colombia (336 images). Images sit under data/test/<set>/ as PNG with a metadata.csv; per-set balanced CSVs give scientific and common names and full seven-rank taxonomy, and potential-sets/ holds the filtered LILA image URL and label tables plus the notebooks that sampled them. Suited to evaluating CLIP-style models (CLIP, SigLIP, BioCLIP, BioCLIP 2) on wildlife camera trap images. License CDLA-Permissive-1.0, the same as the LILA source images. Cite the dataset (Campolongo, Gu, Zhang 2026, DOI 10.57967/hf/10148), BioCLIP 2 and each source dataset.","displayTitle":"IDLE-OO Camera Traps: Balanced Five-Set Species Classification Benchmark from LILA BC Camera Trap Images (Imageomics, BioCLIP 2)","doi":"10.57967/hf/10148","huggingFaceUrl":"https://huggingface.co/datasets/imageomics/IDLE-OO-Camera-Traps","id":"d3a6279f-a6d4-5665-b9f5-8c1e4f64c287","license":"CDLA-Permissive-1.0","licenseEvidenceUrl":"https://huggingface.co/datasets/imageomics/IDLE-OO-Camera-Traps","objectCountEstimate":2613,"probeStatus":"ok","referenceUrl":"https://huggingface.co/datasets/imageomics/IDLE-OO-Camera-Traps","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"imageomics/IDLE-OO-Camera-Traps\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"imageomics/IDLE-OO-Camera-Traps\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=imageomics, size_categories:1K<n<10K, format:imagefolder, downloads=2046"],"format":"mixed","generatedAt":"2026-11-21T20:48:46.187Z","layout":{"example":"ENA24-balanced.csv","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"url_gcp,url_aws,url_azure,image_id,sequence_id,location_id,frame_num,original_label,scientific_name,common_name,kingdom,phylum,cls,order,family,genus,species,filepath,uuid,hashsum_original,hashsum_resized\nhttps://storage.googleapis.com/public-datasets-lila/ena24/images/1001.jpg,http://us-west-2.opendata.source.coop.s3.amazonaws.com/agentmorris/lila-wildlife/ena24/images/1001.jpg,https://lilawildlife.blob.core.windows.net/lila-wildlife/ena24/images/1001.jpg,ENA24 : 1001,ENA24 : unknown,ENA24,-1,eastern chipm","encoding":"utf8","key":"ENA24-balanced.csv","truncatedAt":512}],"sampleObjects":["ENA24-balanced.csv","data/test/ENA24/00170382-1e3d-4d46-9243-ac78136922ba.png","data/test/ENA24/006ff528-6ab8-4679-b0e1-b4abed31560d.png","data/test/ENA24/00cc4c84-b11c-4423-a8f9-1fd2a8c209c9.png","data/test/ENA24/00da5609-7eae-48ea-a77f-6fd3712a4d94.png"],"sidecars":[{"bytes":"---\nlanguage:\n- en\n- la\npretty_name: IDLE-OO Camera Traps\ntags:\n- biology\n- image\n- imageomics\n- animals\n- CV\n- balanced\n- camera traps\n- mammals\n- birds\n- reptiles\n- amphibians\n- lions\n- rodents\n- frogs\n- toads\n- island\n- desert\n- ohio\nsize_categories:\n- 1K<n<10K\ntask_categories:\n- image-classification\n- zero-shot-classification\ndescription: \"5-dataset benchmark of camera trap images from the Labeled Information Library of Alexandria: Biology and Conservation (LILA BC) with a total of 2,586 images for species classification. Each of the 5 benchmarks is **balanced** to have the same number of images for each species within it (between 310 and 1120 images), representing between 16 and 39 species.\"\n---\n\n# Dataset Card for IDLE-OO Camera Traps\n\nIDLE-OO Camera Traps is a 5-dataset benchmark of camera trap images from the [Labeled Information Library of Alexandria: Biology and Conservation (LILA BC)](https://lila.science) with a total of 2,586 images for species classification. Each of the 5 benchmarks is **balanced** to have the same number of images for each species within it (between 310 and 1120 images), representing between 16 and 39 species.\n\n### Dataset Description\n\n- **Curated by:** Elizabeth Campolongo, Jianyang Gu, and Net Zhang\n- **Homepage:** [BioCLIP 2 Project Page](https://imageomics.github.io/bioclip-2/)\n- **Paper:** [BioCLIP 2: Emergent Properties from Scaling Hierarchical Contrastive Learning](https://doi.org/10.48550/arXiv.2505.23883)\n\n### Supported Tasks and Leaderboards\nImage classification, particularly for species classification in camera trap images.\n\n<img src=\"https://cdn-uploads.huggingface.co/production/uploads/642dbba7c3a53b9e1d6e8726/u1zSM8gIq9Nbxa4h_knpF.png\" alt=\"Table 7 from BioCLIP 2 paper indicating best and second best results for the various camera trap components with zero-,one-, and five-shot classification for CLIP, SigLIP, BioTrove-CLIP, BioCLIP, and BioCLIP 2. Supervised-IN21K and DINOv3 results are included for one-and five-shot classification. BioCLIP 2 has the overall best performance in all categories.\" style=\"max-width: 800px; width: 100%; height: auto;\" />\n\n### Languages\nEnglish, Latin\n\n## Dataset Structure\n\n```\n/dataset/\n    desert-lion-balanced.csv\n    ENA24-balanced.csv\n    island-balanced.csv\n    ohio-small-animals-balanced.csv\n    orinoquia-balanced.csv\n    data/test/\n        desert-lion/\n            <image 1>\n            <image 2>\n            ...\n            <image 352>\n        ENA24/\n            <image 1>\n            <image 2>\n            ...\n            <image 1120>\n        island/\n            <image 1>\n            <image 2>\n            ...\n            <image 310>\n        ohio-small-animals/\n            <image 1>\n            <image 2>\n            ...\n            <image 468>\n        orinoquia/\n            <image 1>\n            <image 2>\n            ...\n            <image 336>\n        metadata.csv\n    notebooks/\n        lilabc_CT.ipynb\n        lilabc_CT.py\n        lilabc_test-<dataset_name>.ipynb\n        lilabc_test-filter.ipynb\n        lilabc_test-filter.py\n    potential-sets/\n        lila-taxonomy-mapping_release.csv\n        lila_image_urls_and_labels.csv\n        <dataset_name>_image_urls_and_labels.csv\n```\n\n### Data Instances\n\n**potential-sets/lila_image_urls_and_labels.csv:** Reduced down to the datasets of interest listed below (from [potential-sets/lila_image_urls_and_labels.csv](https://huggingface.co/datasets/imageomics/IDLE-OO-Camera-Traps/blob/37b93ddf25c63bc30d8488ef78c1a53b9c4a3115/data/potential-sets/lila_image_urls_and_labels.csv) (sha256:3fdf87ceea75f8720208a95350c3c70831a6c1c745a92bb68c7f2c3239e4c455)); all those with `original_label` \"empty\" or null `scientific_name` (these had non-taxa labels) were removed. \nAdditionally, we added a `multi_species` column (boolean to indicate multiple species are present in the image--it gets listed once for each species in the image) and a count of how many different species are in each of those images (`num_species` column).\nThis was then subdivided into CSVs for each of the target datasets (`potential-sets/<dataset_name>","encoding":"utf8","key":"README.md","name":"README.md","size":17864,"truncated":true}],"sourceObjectKey":"ENA24-balanced.csv","starterCell":"import pandas as pd\ndf = pd.read_csv(\"hf://datasets/imageomics/IDLE-OO-Camera-Traps/ENA24-balanced.csv\")\ndf.head()","starterSamplePath":"ENA24-balanced.csv","totalObjects":2613},"summary":"A species classification benchmark of 2,586 camera trap images drawn from five datasets in the Labeled Information Library of Alexandria: Biology and Conservation (LILA BC), each…","title":"IDLE-OO Camera Traps: Balanced Five-Set Species Classification Benchmark from LILA BC Camera Trap Images (Imageomics, BioCLIP 2)","updatedAt":"2026-11-22T00:47:59.500Z","visibility":"public","websiteUrl":"https://imageomics.github.io/bioclip-2/"}}