{"url":"/dataset/uvo","name":"UVO","full_name":"Unidentified Video Objects: A Benchmark for Dense, Open-World Segmentation","description_markdown":"UVO is a new benchmark for open-world class-agnostic object segmentation in videos. Besides shifting the problem focus to the open-world setup, UVO is significantly larger, providing approximately 8 times more videos compared with [DAVIS](/dataset/davis), and 7 times more mask (instance) annotations per video compared with [YouTube-VOS](/dataset/youtube-vos) and [YouTube-VIS](/dataset/youtubevis). UVO is also more challenging as it includes many videos with crowded scenes and complex background motions. Some highlights of the dataset include:\r\n\r\n- High quality instance masks densely annotated at 30 fps on 1024 YouTube videos and 1fps on 10337 videos from Kinetics dataset\r\n\r\n- Open-world: annotating all objects in each video, 13.5 objects per video on average\r\n\r\n- Diverse object categories: 57% of objects are not covered by COCO categories","description_withheld":null,"homepage":"https://sites.google.com/view/unidentified-video-object/home?authuser=0","introduced_date":null,"introduced_date_note":null,"introduced_by":{"paper":"/paper/unidentified-video-objects-a-benchmark-for","title":"Unidentified Video Objects: A Benchmark for Dense, Open-World Segmentation","first_author":"Weiyao Wang","url":null},"license":null,"modalities":[{"name":"Images","url":"/datasets/modality/images"},{"name":"Videos","url":"/datasets/modality/videos"},{"name":"RGB Video","url":"/datasets/modality/rgb-video"}],"tasks":[{"name":"Object Detection","url":"/task/object-detection","datasets_with_task":"/datasets/task/object-detection"},{"name":"Instance Segmentation","url":"/task/instance-segmentation","datasets_with_task":"/datasets/task/instance-segmentation"},{"name":"Object Localization","url":"/task/object-localization","datasets_with_task":"/datasets/task/object-localization"},{"name":"Unsupervised Object Detection","url":"/task/unsupervised-object-detection","datasets_with_task":"/datasets/task/unsupervised-object-detection"},{"name":"Video Instance Segmentation","url":"/task/video-instance-segmentation","datasets_with_task":"/datasets/task/video-instance-segmentation"},{"name":"Open World Object Detection","url":"/task/open-world-object-detection","datasets_with_task":"/datasets/task/open-world-object-detection"},{"name":"Unsupervised Instance Segmentation","url":"/task/unsupervised-instance-segmentation","datasets_with_task":"/datasets/task/unsupervised-instance-segmentation"},{"name":"Open-World Instance Segmentation","url":"/task/open-world-instance-segmentation","datasets_with_task":"/datasets/task/open-world-instance-segmentation"},{"name":"Object Discovery","url":"/task/object-discovery","datasets_with_task":"/datasets/task/object-discovery"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["UVO"],"data_loaders":[{"repo":"https://github.com/dulucas/uvo_challenge","url":"https://github.com/dulucas/uvo_challenge","frameworks":["pytorch"]}],"num_papers_in_archive":27,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/open-world-instance-segmentation-on-uvo","task":"Open-World Instance Segmentation","dataset_variant":"UVO","rows":2,"metrics":["ARmask"],"first_row_in_archive_order":{"model":"GLEE-Pro","paper":"/paper/general-object-foundation-model-for-images","metrics":{"ARmask":"72.6"},"code_links":[{"title":"FoundationVision/GLEE","url":"https://github.com/FoundationVision/GLEE"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/unsupervised-instance-segmentation-on-uvo","task":"Unsupervised Instance Segmentation","dataset_variant":"UVO","rows":1,"metrics":["AP","AP50","AP75"],"first_row_in_archive_order":{"model":"CutLER (Cascade+DINO)","paper":"/paper/cut-and-learn-for-unsupervised-object","metrics":{"AP":"10.1","AP50":"22.8","AP75":"8"},"code_links":[{"title":"facebookresearch/cutler","url":"https://github.com/facebookresearch/cutler"},{"title":"u2seg/u2seg","url":"https://github.com/u2seg/u2seg"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/general-object-foundation-model-for-images","title":"General Object Foundation Model for Images and Videos at Scale","date":"2023-12-14","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":13,"samples_ran":8,"samples_unverified":5,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/open-vocabulary-panoptic-segmentation-with-1","title":"Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models","date":"2023-03-08","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/cut-and-learn-for-unsupervised-object","title":"Cut and Learn for Unsupervised Object Detection and Instance Segmentation","date":"2023-01-26","rows_on_this_dataset":1,"code_links":2,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":1,"samples_harvested":13,"samples_ran":8,"samples_unverified":5,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}