{"url":"/sota/referring-video-object-segmentation-on-mevis","task":{"name":"Referring Video Object Segmentation","url":"/task/referring-video-object-segmentation","note":null},"dataset":{"name":"MeViS","url":"/dataset/mevis"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Referring video object segmentation aims at segmenting an object in video with language expressions. Unlike the previous video object segmentation, the task exploits a different type of supervision, language expressions, to identify and segment an object referred by the given language expressions in a video.","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["J&F","J","F"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"J&F":null,"J":null,"F":null}},"counts":{"rows":16,"rows_with_code":15,"rows_with_paper_page":16,"rows_dated":15,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"MPG-SAM 2","metrics":{"F":"56.7","J":"50.7","J&F":"53.7"},"uses_additional_data":false,"paper_date":"2025-01-23","paper":"/paper/mpg-sam-2-adapting-sam-2-with-mask-priors-and","paper_url":"https://arxiv.org/abs/2501.13667v1","paper_title":"MPG-SAM 2: Adapting SAM 2 with Mask Priors and Global Context for Referring Video Object Segmentation","code":"https://github.com/rongfu-dsb/MPG-SAM2","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":11,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"FindTrack","metrics":{"F":"55.9","J":"50.5","J&F":"53.2"},"uses_additional_data":false,"paper_date":"2025-03-05","paper":"/paper/find-first-track-next-decoupling","paper_url":"https://arxiv.org/abs/2503.03492v1","paper_title":"Find First, Track Next: Decoupling Identification and Propagation in Referring Video Object Segmentation","code":"https://github.com/suhwan-cho/FindTrack","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"GLUS","metrics":{"F":"54.2","J":"48.5","J&F":"51.3"},"uses_additional_data":false,"paper_date":"2025-04-10","paper":"/paper/glus-global-local-reasoning-unified-into-a","paper_url":"https://arxiv.org/abs/2504.07962v1","paper_title":"GLUS: Global-Local Reasoning Unified into A Single Large Language Model for Video Segmentation","code":"https://github.com/GLUS-video/GLUS","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":9,"n_samples":12,"n_pointer_only_licence":12}},{"rank_in_archive_order":4,"model":"VRS-HQ (Chat-UniVi-13B)","metrics":{"F":"53.7","J":"48","J&F":"50.9"},"uses_additional_data":false,"paper_date":"2025-01-15","paper":"/paper/the-devil-is-in-temporal-token-high-quality","paper_url":"https://arxiv.org/abs/2501.08549v1","paper_title":"The Devil is in Temporal Token: High Quality Video Reasoning Segmentation","code":"https://github.com/sitonggong/vrs-hq","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"ReferDINO (Swin-B)","metrics":{"F":"53.9","J":"44.7","J&F":"49.3"},"uses_additional_data":false,"paper_date":"2025-01-24","paper":"/paper/referdino-referring-video-object-segmentation","paper_url":"https://arxiv.org/abs/2501.14607v1","paper_title":"ReferDINO: Referring Video Object Segmentation with Visual Grounding Foundations","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"SAMWISE","metrics":{"F":"51.2","J":"45.4","J&F":"48.3"},"uses_additional_data":false,"paper_date":"2024-11-26","paper":"/paper/samwise-infusing-wisdom-in-sam2-for-text","paper_url":"https://arxiv.org/abs/2411.17646v2","paper_title":"SAMWISE: Infusing Wisdom in SAM2 for Text-Driven Video Segmentation","code":"https://github.com/claudiacuttano/samwise","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"DsHmp + MTCM","metrics":{"F":"51.1","J":"44.1","J&F":"47.6"},"uses_additional_data":false,"paper_date":"2025-01-09","paper":"/paper/multi-context-temporal-consistent-modeling","paper_url":"https://arxiv.org/abs/2501.04939v2","paper_title":"Multi-Context Temporal Consistent Modeling for Referring Video Object Segmentation","code":"https://github.com/choi58/mtcm","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"DsHmp","metrics":{"F":"49.8","J":"43","J&F":"46.4"},"uses_additional_data":false,"paper_date":"2024-04-04","paper":"/paper/decoupling-static-and-hierarchical-motion","paper_url":"https://arxiv.org/abs/2404.03645v1","paper_title":"Decoupling Static and Hierarchical Motion Perception for Referring Video Segmentation","code":"https://github.com/heshuting555/dshmp","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":5,"n_samples":7,"n_pointer_only_licence":7}},{"rank_in_archive_order":9,"model":"HTR","metrics":{"F":"45.5","J":"39.9","J&F":"42.7"},"uses_additional_data":false,"paper_date":"2024-03-28","paper":"/paper/towards-temporally-consistent-referring-video","paper_url":"https://arxiv.org/abs/2403.19407v2","paper_title":"Temporally Consistent Referring Video Object Segmentation with Hybrid Memory","code":"https://github.com/bo-miao/HTR","n_code_links":1,"syntology":{"n_ran":14,"n_unverified":1,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"LMPM","metrics":{"F":"40.2","J":"34.2","J&F":"37.2"},"uses_additional_data":false,"paper_date":"2023-08-16","paper":"/paper/mevis-a-large-scale-benchmark-for-video","paper_url":"https://arxiv.org/abs/2308.08544v1","paper_title":"MeViS: A Large-scale Benchmark for Video Segmentation with Motion Expressions","code":"https://github.com/henghuiding/MeViS","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":5,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"VLT+TC","metrics":{"F":"37.3","J":"33.6","J&F":"35.5"},"uses_additional_data":false,"paper_date":"2022-10-28","paper":"/paper/vlt-vision-language-transformer-and-query","paper_url":"https://arxiv.org/abs/2210.15871v1","paper_title":"VLT: Vision-Language Transformer and Query Generation for Referring Segmentation","code":"https://github.com/henghuiding/Vision-Language-Transformer","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"InternVideo2.5","metrics":{"J&F":"32"},"uses_additional_data":false,"paper_date":"2025-01-21","paper":"/paper/internvideo2-5-empowering-video-mllms-with","paper_url":"https://arxiv.org/abs/2501.12386v2","paper_title":"InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling","code":"https://github.com/opengvlab/internvideo","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"ReferFormer","metrics":{"F":"32.2","J":"29.8","J&F":"31.0"},"uses_additional_data":false,"paper_date":"2022-01-03","paper":"/paper/language-as-queries-for-referring-video","paper_url":"https://arxiv.org/abs/2201.00487v2","paper_title":"Language as Queries for Referring Video Object Segmentation","code":"https://github.com/wjn922/referformer","n_code_links":1,"syntology":{"n_ran":7,"n_unverified":1,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":14,"model":"MTTR","metrics":{"F":"31.2","J":"28.8","J&F":"30.0"},"uses_additional_data":false,"paper_date":"2021-11-29","paper":"/paper/end-to-end-referring-video-object","paper_url":"https://arxiv.org/abs/2111.14821v2","paper_title":"End-to-End Referring Video Object Segmentation with Multimodal Transformers","code":"https://github.com/mttr2021/MTTR","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":5,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"LBDT","metrics":{"F":"30.8","J":"27.8","J&F":"29.3"},"uses_additional_data":false,"paper_date":"2022-06-08","paper":"/paper/language-bridged-spatial-temporal-interaction-1","paper_url":"https://arxiv.org/abs/2206.03789v1","paper_title":"Language-Bridged Spatial-Temporal Interaction for Referring Video Object Segmentation","code":"https://github.com/dzh19990407/lbdt","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":2,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"URVOS","metrics":{"F":"29.9","J":"25.7","J&F":"27.8"},"uses_additional_data":false,"paper_date":null,"paper":"/paper/urvos-unified-referring-video-object","paper_url":"https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2327_ECCV_2020_paper.php","paper_title":"URVOS: Unified Referring Video Object Segmentation Network with a Large-Scale Benchmark","code":"https://github.com/skynbe/Refer-Youtube-VOS","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,795 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":9,"rows_with_any_sample_ran":7,"distinct_papers_with_graph_line":9,"distinct_papers_with_any_sample_ran":7,"samples_over_distinct_papers":{"n_ran":43,"n_unverified":45,"n_samples":88,"n_pointer_only_licence":27,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":43,"n_unverified":45,"n_samples":88,"n_pointer_only_licence":27,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}