{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/omnivore-a-single-model-for-many-visual","title":"Omnivore: A Single Model for Many Visual Modalities","arxiv_id":"2201.08377","date":"2022-01-20","proceeding":"CVPR 2022 1","authors":["Rohit Girdhar","Mannat Singh","Nikhila Ravi","Laurens van der Maaten","Armand Joulin","Ishan Misra"],"abstract":"Prior work has studied different visual modalities in isolation and developed separate architectures for recognition of images, videos, and 3D data. Instead, in this paper, we propose a single model which excels at classifying images, videos, and single-view 3D data using exactly the same model parameters. Our 'Omnivore' model leverages the flexibility of transformer-based architectures and is trained jointly on classification tasks from different modalities. Omnivore is simple to train, uses off-the-shelf standard datasets, and performs at-par or better than modality-specific models of the same size. A single Omnivore model obtains 86.0% on ImageNet, 84.1% on Kinetics, and 67.1% on SUN RGB-D. After finetuning, our models outperform prior work on a variety of vision tasks and generalize across modalities. Omnivore's shared visual representation naturally enables cross-modal recognition without access to correspondences between modalities. We hope our results motivate researchers to model visual modalities together.","url_abs":"https://arxiv.org/abs/2201.08377v2","url_pdf":"https://arxiv.org/pdf/2201.08377v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"omnivore-a-single-model-for-many-visual","repo_url":"https://github.com/facebookresearch/omnivore","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"omnivore-a-single-model-for-many-visual","repo_url":"https://github.com/towhee-io/towhee","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"action-classification","task_name":"Action Classification"},{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"scene-recognition","task_name":"Scene Recognition"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"model","task_name":"model"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/action-classification-on-kinetics-400","task":"Action Classification","dataset":"Kinetics-400","model":"OMNIVORE (Swin-L)","rank_in_archive_order":62,"of":207,"metrics":{"Acc@1":"84.1","Acc@5":"96.1"},"uses_additional_data":false},{"leaderboard":"/sota/action-classification-on-kinetics-400","task":"Action Classification","dataset":"Kinetics-400","model":"OMNIVORE (Swin-B)","rank_in_archive_order":63,"of":207,"metrics":{"Acc@1":"84.0","Acc@5":"96.2"},"uses_additional_data":false},{"leaderboard":"/sota/action-recognition-on-epic-kitchens-100","task":"Action Recognition","dataset":"EPIC-KITCHENS-100","model":"OMNIVORE (Swin-B, finetuned)","rank_in_archive_order":9,"of":32,"metrics":{"Action@1":"49.9","Noun@1":"61.7","Verb@1":"69.5"},"uses_additional_data":true},{"leaderboard":"/sota/action-recognition-in-videos-on-something","task":"Action Recognition","dataset":"Something-Something V2","model":"OMNIVORE (Swin-B,  IN-21K+ Kinetics400 pretrain)","rank_in_archive_order":29,"of":123,"metrics":{"Top-1 Accuracy":"71.4","Top-5 Accuracy":"93.5"},"uses_additional_data":true},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"Omnivore (Swin-L)","rank_in_archive_order":177,"of":1060,"metrics":{"Top 1 Accuracy":"86.0%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"Omnivore (Swin-B)","rank_in_archive_order":239,"of":1060,"metrics":{"Top 1 Accuracy":"85.3%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-inaturalist-2018","task":"Image Classification","dataset":"iNaturalist 2018","model":"OMNIVORE (Swin-L)","rank_in_archive_order":11,"of":60,"metrics":{"Top-1 Accuracy":"84.1%"},"uses_additional_data":true},{"leaderboard":"/sota/scene-recognition-on-sun-rgbd","task":"Scene Recognition","dataset":"SUN-RGBD","model":"OMNIVORE (Swin-B)","rank_in_archive_order":1,"of":2,"metrics":{"Accuracy (%)":"67.2"},"uses_additional_data":true},{"leaderboard":"/sota/semantic-segmentation-on-nyu-depth-v2","task":"Semantic Segmentation","dataset":"NYU Depth v2","model":"OMNIVORE (Swin-L, finetuned)","rank_in_archive_order":19,"of":121,"metrics":{"Mean IoU":"56.8%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-nyu-depth-v2","task":"Semantic Segmentation","dataset":"NYU Depth v2","model":"OMNIVORE (Swin-B, finetuned)","rank_in_archive_order":28,"of":121,"metrics":{"Mean IoU":"55.1%"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2201.08377","atlas_url":"https://app.syntology.ai/?focus=2201.08377","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2201.08377"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/omnivore","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/towhee-io/towhee","reach":null}],"summary":{"ran":2},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1},"listed":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"9b5687fccc4bf734","entry":"OmnivoreModel","repo":"facebookresearch/omnivore","repo_kind":"official","path":"omnivore/models/omnivore_model.py","file_url":"https://github.com/facebookresearch/omnivore/blob/HEAD/omnivore/models/omnivore_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"9b5687fccc4bf734"}},{"code_sha256_prefix":"d92bf522f8f7f64f","entry":"OmnivoreModel","repo":"towhee-io/towhee","repo_kind":"listed","path":"towhee/models/omnivore/omnivore.py","file_url":"https://github.com/towhee-io/towhee/blob/HEAD/towhee/models/omnivore/omnivore.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d92bf522f8f7f64f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}