{"url":"/sota/video-instance-segmentation-on-youtube-vis-1","task":{"name":"Video Instance Segmentation","url":"/task/video-instance-segmentation","note":null},"dataset":{"name":"YouTube-VIS validation","url":"/dataset/youtubevis"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"The goal of video instance segmentation is simultaneous detection, segmentation and tracking of instances in videos. In words, it is the first time that the image instance segmentation problem is extended to the video domain.\r\n\r\nTo facilitate research on this new task, a large-scale benchmark called YouTube-VIS, which consists of 2,883 high-resolution YouTube videos, a 40-category label set and 131k high-quality instance masks is built.","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":["mask AP","AP50","AP75","AR1","AR10"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"mask AP":"higher","AP50":null,"AP75":null,"AR1":null,"AR10":null}},"counts":{"rows":44,"rows_with_code":41,"rows_with_paper_page":44,"rows_dated":44,"rows_using_additional_data":10},"rows":[{"rank_in_archive_order":1,"model":"CAVIS(ViT-L, Online)","metrics":{"AP50":"89.3","AP75":"76.2","AR1":"58.3","AR10":"73.6","mask AP":"68.9"},"uses_additional_data":true,"paper_date":"2024-07-03","paper":"/paper/context-aware-video-instance-segmentation","paper_url":"https://arxiv.org/abs/2407.03010v1","paper_title":"Context-Aware Video Instance Segmentation","code":"https://github.com/Seung-Hun-Lee/CAVIS","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"DVIS++(ViT-L, Online)","metrics":{"AP50":"88.8","AP75":"75.3","AR1":"57.9","AR10":"73.7","mask AP":"67.7"},"uses_additional_data":true,"paper_date":"2023-12-20","paper":"/paper/dvis-improved-decoupled-framework-for","paper_url":"https://arxiv.org/abs/2312.13305v1","paper_title":"DVIS++: Improved Decoupled Framework for Universal Video Segmentation","code":"https://github.com/zhang-tao-whu/DVIS_Plus","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"DVIS","metrics":{"AP50":"88.0","AP75":"72.7","AR1":"56.5","AR10":"70.3","mask AP":"64.9"},"uses_additional_data":true,"paper_date":"2023-06-06","paper":"/paper/dvis-decoupled-video-instance-segmentation","paper_url":"https://arxiv.org/abs/2306.03413v3","paper_title":"DVIS: Decoupled Video Instance Segmentation Framework","code":"https://github.com/zhang-tao-whu/DVIS","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"Tube-Link","metrics":{"AP50":"86.6","AP75":"71.3","AR1":"55.9","AR10":"69.1","mask AP":"64.6"},"uses_additional_data":false,"paper_date":"2023-03-22","paper":"/paper/tube-link-a-flexible-cross-tube-baseline-for","paper_url":"https://arxiv.org/abs/2303.12782v3","paper_title":"Tube-Link: A Flexible Cross Tube Framework for Universal Video Segmentation","code":"https://github.com/lxtgh/tube-link","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"MinVIS (Swin-L)","metrics":{"AP50":"83.3","AP75":"68.6","AR1":"54.8","AR10":"66.6","mask AP":"61.6"},"uses_additional_data":false,"paper_date":"2022-08-03","paper":"/paper/minvis-a-minimal-video-instance-segmentation","paper_url":"https://arxiv.org/abs/2208.02245v1","paper_title":"MinVIS: A Minimal Video Instance Segmentation Framework without Video-based Training","code":"https://github.com/nvlabs/minvis","n_code_links":2,"syntology":null},{"rank_in_archive_order":6,"model":"Mask2Former (Swin-L)","metrics":{"AP50":"84.4","AP75":"67.0","mask AP":"60.4"},"uses_additional_data":false,"paper_date":"2021-12-20","paper":"/paper/mask2former-for-video-instance-segmentation","paper_url":"https://arxiv.org/abs/2112.10764v1","paper_title":"Mask2Former for Video Instance Segmentation","code":"https://github.com/huggingface/transformers","n_code_links":6,"syntology":{"n_ran":2,"n_unverified":5,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"UniVS(Swin-L)","metrics":{"AP50":"82.1","AP75":"65.3","AR1":"54.7","AR10":"66.8","mask AP":"60.0"},"uses_additional_data":true,"paper_date":"2024-02-28","paper":"/paper/univs-unified-and-universal-video","paper_url":"https://arxiv.org/abs/2402.18115v2","paper_title":"UniVS: Unified and Universal Video Segmentation with Prompts as Queries","code":"https://github.com/minghanli/univs","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":2,"n_samples":14,"n_pointer_only_licence":14}},{"rank_in_archive_order":8,"model":"MDQE(Swin-L)","metrics":{"AP50":"84.9","AP75":"67.3","AR1":"53.5","AR10":"65.0","mask AP":"59.9"},"uses_additional_data":false,"paper_date":"2023-03-25","paper":"/paper/mdqe-mining-discriminative-query-embeddings","paper_url":"https://arxiv.org/abs/2303.14395v1","paper_title":"MDQE: Mining Discriminative Query Embeddings to Segment Occluded Instances on Challenging Videos","code":"https://github.com/minghanli/mdqe_cvpr2023","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":1,"n_samples":4,"n_pointer_only_licence":4}},{"rank_in_archive_order":9,"model":"SeqFormer (Swin-L)","metrics":{"AP50":"82.1","AP75":"66.4","AR1":"51.7","AR10":"64.4","mask AP":"59.3"},"uses_additional_data":true,"paper_date":"2021-12-15","paper":"/paper/seqformer-a-frustratingly-simple-model-for","paper_url":"https://arxiv.org/abs/2112.08275v2","paper_title":"SeqFormer: Sequential Transformer for Video Instance Segmentation","code":"https://github.com/wjf5203/vnext","n_code_links":2,"syntology":null},{"rank_in_archive_order":10,"model":"DeVIS (Swin-L)","metrics":{"AP50":"80.8","AP75":"66.3","AR1":"50.8","AR10":"61.0","mask AP":"57.1"},"uses_additional_data":false,"paper_date":"2022-07-22","paper":"/paper/devis-making-deformable-transformers-work-for","paper_url":"https://arxiv.org/abs/2207.11103v1","paper_title":"DeVIS: Making Deformable Transformers Work for Video Instance Segmentation","code":"https://github.com/acaelles97/devis","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":2,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":11,"model":"InstanceFormer(Swin-L)","metrics":{"AP50":"78.0","AP75":"64.2","AR1":"50.9","AR10":"61.6","mask AP":"56.3"},"uses_additional_data":true,"paper_date":"2022-08-22","paper":"/paper/instanceformer-an-online-video-instance","paper_url":"https://arxiv.org/abs/2208.10547v1","paper_title":"InstanceFormer: An Online Video Instance Segmentation Framework","code":"https://github.com/rajatkoner08/instanceformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"TCIS (Swin-S)","metrics":{"AP50":"76.6","AP75":"65.6","AR1":"47","AR10":"57.9","mask AP":"54.3"},"uses_additional_data":false,"paper_date":"2021-06-12","paper":"/paper/1st-place-solution-for-youtubevos-challenge","paper_url":"https://arxiv.org/abs/2106.06649v2","paper_title":"1st Place Solution for YouTubeVOS Challenge 2021:Video Instance Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"Video K-Net (Swin-Base)","metrics":{"AP50":"79.0","AP75":"59.6","AR1":"49.7","AR10":"59.9","mask AP":"54.1"},"uses_additional_data":false,"paper_date":"2022-04-10","paper":"/paper/video-k-net-a-simple-strong-and-unified","paper_url":"https://arxiv.org/abs/2204.04656v2","paper_title":"Video K-Net: A Simple, Strong, and Unified Baseline for Video Segmentation","code":"https://github.com/lxtgh/video-k-net","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"NOVIS (ResNet-50)","metrics":{"AP50":"75.7","AP75":"56.9","AR1":"50.3","AR10":"60.6","mask AP":"52.8"},"uses_additional_data":true,"paper_date":"2023-08-29","paper":"/paper/novis-a-case-for-end-to-end-near-online-video","paper_url":"https://arxiv.org/abs/2308.15266v2","paper_title":"NOVIS: A Case for End-to-End Near-Online Video Instance Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"IDOL (ResNet-50)","metrics":{"AP50":"74","AP75":"52.9","AR1":"47.7","AR10":"58.7","mask AP":"49.5"},"uses_additional_data":false,"paper_date":"2022-07-21","paper":"/paper/in-defense-of-online-models-for-video","paper_url":"https://arxiv.org/abs/2207.10661v1","paper_title":"In Defense of Online Models for Video Instance Segmentation","code":"https://github.com/wjf5203/vnext","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":2,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"Mask2Former (ResNet-101)","metrics":{"AP50":"72.8","AP75":"54.2","mask AP":"49.2"},"uses_additional_data":false,"paper_date":"2021-12-20","paper":"/paper/mask2former-for-video-instance-segmentation","paper_url":"https://arxiv.org/abs/2112.10764v1","paper_title":"Mask2Former for Video Instance Segmentation","code":"https://github.com/huggingface/transformers","n_code_links":6,"syntology":{"n_ran":2,"n_unverified":5,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"SeqFormer (ResNet-101)","metrics":{"AP50":"71.1","AP75":"55.7","AR1":"46.8","AR10":"56.9","mask AP":"49.0"},"uses_additional_data":true,"paper_date":"2021-12-15","paper":"/paper/seqformer-a-frustratingly-simple-model-for","paper_url":"https://arxiv.org/abs/2112.08275v2","paper_title":"SeqFormer: Sequential Transformer for Video Instance Segmentation","code":"https://github.com/wjf5203/vnext","n_code_links":2,"syntology":null},{"rank_in_archive_order":18,"model":"MSN","metrics":{"AP50":"69.4","AP75":"54.9","AR1":"40.1","AR10":"55.0","mask AP":"48.8"},"uses_additional_data":false,"paper_date":"2021-06-19","paper":"/paper/msn-efficient-online-mask-selection-network","paper_url":"https://arxiv.org/abs/2106.10452v1","paper_title":"MSN: Efficient Online Mask Selection Network for Video Instance Segmentation","code":"https://github.com/SHI-Labs/Mask-Selection-Networks","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"SeqFormer (ResNet-50)","metrics":{"AP50":"69.8","AP75":"51.8","AR1":"45.5","AR10":"54.8","mask AP":"47.4"},"uses_additional_data":true,"paper_date":"2021-12-15","paper":"/paper/seqformer-a-frustratingly-simple-model-for","paper_url":"https://arxiv.org/abs/2112.08275v2","paper_title":"SeqFormer: Sequential Transformer for Video Instance Segmentation","code":"https://github.com/wjf5203/vnext","n_code_links":2,"syntology":null},{"rank_in_archive_order":20,"model":"Mask2Former (ResNet-50)","metrics":{"AP50":"68.0","AP75":"50.0","mask AP":"46.4"},"uses_additional_data":false,"paper_date":"2021-12-20","paper":"/paper/mask2former-for-video-instance-segmentation","paper_url":"https://arxiv.org/abs/2112.10764v1","paper_title":"Mask2Former for Video Instance Segmentation","code":"https://github.com/huggingface/transformers","n_code_links":6,"syntology":{"n_ran":2,"n_unverified":5,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":21,"model":"InstanceFormer(ResNet-50)","metrics":{"AP50":"68.6","AP75":"49.6","AR1":"42.1","AR10":"53.5","mask AP":"45.6"},"uses_additional_data":true,"paper_date":"2022-08-22","paper":"/paper/instanceformer-an-online-video-instance","paper_url":"https://arxiv.org/abs/2208.10547v1","paper_title":"InstanceFormer: An Online Video Instance Segmentation Framework","code":"https://github.com/rajatkoner08/instanceformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"SeqFormer (ResNet-50)","metrics":{"AP50":"66.9","AP75":"50.5","AR1":"45.6","AR10":"54.6","mask AP":"45.1"},"uses_additional_data":false,"paper_date":"2021-12-15","paper":"/paper/seqformer-a-frustratingly-simple-model-for","paper_url":"https://arxiv.org/abs/2112.08275v2","paper_title":"SeqFormer: Sequential Transformer for Video Instance Segmentation","code":"https://github.com/wjf5203/vnext","n_code_links":2,"syntology":null},{"rank_in_archive_order":23,"model":"DeVIS (ResNet-50)","metrics":{"AP50":"66.7","AP75":"48.6","AR1":"42.4","AR10":"51.6","mask AP":"44.4"},"uses_additional_data":false,"paper_date":"2022-07-22","paper":"/paper/devis-making-deformable-transformers-work-for","paper_url":"https://arxiv.org/abs/2207.11103v1","paper_title":"DeVIS: Making Deformable Transformers Work for Video Instance Segmentation","code":"https://github.com/acaelles97/devis","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":2,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":24,"model":"IFC (ResNet-50)","metrics":{"AP50":"65.8","AP75":"46.8","AR1":"43.8","AR10":"51.2","mask AP":"42.8"},"uses_additional_data":false,"paper_date":"2021-06-07","paper":"/paper/video-instance-segmentation-using-inter-frame","paper_url":"https://arxiv.org/abs/2106.03299v1","paper_title":"Video Instance Segmentation using Inter-Frame Communication Transformers","code":"https://github.com/sukjunhwang/IFC","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":25,"model":"VisTR(ResNet-101)","metrics":{"AP50":"64.0","AP75":"45.0","AR1":"38.3","AR10":"44.9","mask AP":"40.1"},"uses_additional_data":false,"paper_date":"2020-11-30","paper":"/paper/end-to-end-video-instance-segmentation-with","paper_url":"https://arxiv.org/abs/2011.14503v5","paper_title":"End-to-End Video Instance Segmentation with Transformers","code":"https://github.com/Epiphqny/VisTR","n_code_links":2,"syntology":null},{"rank_in_archive_order":26,"model":"VSTAM","metrics":{"mask AP":"39.0"},"uses_additional_data":false,"paper_date":"2022-06-17","paper":"/paper/video-sparse-transformer-with-attention","paper_url":"https://ieeexplore.ieee.org/document/9798833","paper_title":"Video Sparse Transformer With Attention-Guided Memory for Video Object Detection","code":"https://github.com/Malik1998/VSTAM","n_code_links":1,"syntology":null},{"rank_in_archive_order":27,"model":"STMask(R101-DCN-FPN)","metrics":{"AP50":"56.8","AP75":"38.0","AR1":"34.8","AR10":"41.8","mask AP":"36.8"},"uses_additional_data":false,"paper_date":"2021-04-06","paper":"/paper/spatial-feature-calibration-and-temporal","paper_url":"https://arxiv.org/abs/2104.05606v1","paper_title":"Spatial Feature Calibration and Temporal Fusion for Effective One-stage Video Instance Segmentation","code":"https://github.com/MinghanLi/STMask","n_code_links":1,"syntology":null},{"rank_in_archive_order":28,"model":"STC (ResNet-50)","metrics":{"AP50":"57.2","AP75":"38.6","AR1":"36.9","AR10":"44.5","mask AP":"36.7"},"uses_additional_data":false,"paper_date":"2022-02-08","paper":"/paper/stc-spatio-temporal-contrastive-learning-for","paper_url":"https://arxiv.org/abs/2202.03747v2","paper_title":"STC: Spatio-Temporal Contrastive Learning for Video Instance Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":29,"model":"CrossVIS (ResNet-101)","metrics":{"AP50":"57.3","AP75":"39.7","AR1":"36","AR10":"42","mask AP":"36.6"},"uses_additional_data":false,"paper_date":"2021-04-13","paper":"/paper/crossover-learning-for-fast-online-video","paper_url":"https://arxiv.org/abs/2104.05970v1","paper_title":"Crossover Learning for Fast Online Video Instance Segmentation","code":"https://github.com/hustvl/CrossVIS","n_code_links":1,"syntology":null},{"rank_in_archive_order":30,"model":"VisTR(ResNet-50)","metrics":{"AP50":"59.8","AP75":"36.9","AR1":"37.2","AR10":"42.4","mask AP":"36.2"},"uses_additional_data":false,"paper_date":"2020-11-30","paper":"/paper/end-to-end-video-instance-segmentation-with","paper_url":"https://arxiv.org/abs/2011.14503v5","paper_title":"End-to-End Video Instance Segmentation with Transformers","code":"https://github.com/Epiphqny/VisTR","n_code_links":2,"syntology":null},{"rank_in_archive_order":31,"model":"PCAN(ResNet-50)","metrics":{"AP50":"54.9","AP75":"39.4","AR1":"36.3","AR10":"41.6","mask AP":"36.1"},"uses_additional_data":false,"paper_date":"2021-06-22","paper":"/paper/prototypical-cross-attention-networks-for","paper_url":"https://arxiv.org/abs/2106.11958v2","paper_title":"Prototypical Cross-Attention Networks for Multiple Object Tracking and Segmentation","code":"https://github.com/SysCV/pcan","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":5,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":32,"model":"ObjProp (ResNet-50)","metrics":{"AP50":"59.4","AP75":"39.2","AR1":"39.1","AR10":"47.7","mask AP":"36.0"},"uses_additional_data":false,"paper_date":"2021-11-15","paper":"/paper/object-propagation-via-inter-frame-attentions","paper_url":"https://arxiv.org/abs/2111.07529v3","paper_title":"Object Propagation via Inter-Frame Attentions for Temporally Stable Video Instance Segmentation","code":"https://github.com/anirudh-chakravarthy/objprop","n_code_links":1,"syntology":null},{"rank_in_archive_order":33,"model":"CompFeat(ResNet-50)","metrics":{"AP50":"56.0","AP75":"38.6","AR1":"33.1","AR10":"40.3","mask AP":"35.3"},"uses_additional_data":false,"paper_date":"2020-12-07","paper":"/paper/compfeat-comprehensive-feature-aggregation","paper_url":"https://arxiv.org/abs/2012.03400v1","paper_title":"CompFeat: Comprehensive Feature Aggregation for Video Instance Segmentation","code":"https://github.com/SHI-Labs/CompFeat-for-Video-Instance-Segmentation","n_code_links":1,"syntology":null},{"rank_in_archive_order":34,"model":"CSipMask","metrics":{"AP50":"55.6","AP75":"38.1","mask AP":"35.1"},"uses_additional_data":false,"paper_date":"2021-02-02","paper":"/paper/occluded-video-instance-segmentation","paper_url":"https://arxiv.org/abs/2102.01558v6","paper_title":"Occluded Video Instance Segmentation: A Benchmark","code":"https://github.com/haochenheheda/lvvis","n_code_links":2,"syntology":null},{"rank_in_archive_order":35,"model":"STEm-Seg (ResNet-101)","metrics":{"AP50":"55.8","AP75":"37.9","AR1":"34.4","AR10":"41.6","mask AP":"34.6"},"uses_additional_data":false,"paper_date":"2020-03-18","paper":"/paper/stem-seg-spatio-temporal-embeddings-for","paper_url":"https://arxiv.org/abs/2003.08429v4","paper_title":"STEm-Seg: Spatio-temporal Embeddings for Instance Segmentation in Videos","code":"https://github.com/sabarim/STEm-Seg","n_code_links":1,"syntology":null},{"rank_in_archive_order":36,"model":"SipMask (ResNet-50, ms-train, single-scale test)","metrics":{"AP50":"54.1","AP75":"35.8","AR1":"35.4","AR10":"40.1","mask AP":"33.7"},"uses_additional_data":false,"paper_date":"2020-07-29","paper":"/paper/sipmask-spatial-information-preservation-for","paper_url":"https://arxiv.org/abs/2007.14772v1","paper_title":"SipMask: Spatial Information Preservation for Fast Image and Video Instance Segmentation","code":"https://github.com/JialeCao001/SipMask","n_code_links":1,"syntology":null},{"rank_in_archive_order":37,"model":"TraDeS","metrics":{"AP50":"52.6","AP75":"32.8","mask AP":"32.6"},"uses_additional_data":false,"paper_date":"2021-03-16","paper":"/paper/track-to-detect-and-segment-an-online-multi","paper_url":"https://arxiv.org/abs/2103.08808v1","paper_title":"Track to Detect and Segment: An Online Multi-Object Tracker","code":"https://github.com/JialianW/TraDeS","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":0,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":38,"model":"SipMask (ResNet-50, single-scale test)","metrics":{"AP50":"53","AP75":"33.3","AR1":"33.5","AR10":"38.9","mask AP":"32.5"},"uses_additional_data":false,"paper_date":"2020-07-29","paper":"/paper/sipmask-spatial-information-preservation-for","paper_url":"https://arxiv.org/abs/2007.14772v1","paper_title":"SipMask: Spatial Information Preservation for Fast Image and Video Instance Segmentation","code":"https://github.com/JialeCao001/SipMask","n_code_links":1,"syntology":null},{"rank_in_archive_order":39,"model":"CMaskTrack R-CNN","metrics":{"AP50":"52.8","AP75":"34.9","mask AP":"32.1"},"uses_additional_data":false,"paper_date":"2021-02-02","paper":"/paper/occluded-video-instance-segmentation","paper_url":"https://arxiv.org/abs/2102.01558v6","paper_title":"Occluded Video Instance Segmentation: A Benchmark","code":"https://github.com/haochenheheda/lvvis","n_code_links":2,"syntology":null},{"rank_in_archive_order":40,"model":"STEm-Seg (ResNet-50)","metrics":{"AP50":"50.7","AP75":"37.9","AR1":"34.4","AR10":"41.6","mask AP":"30.6"},"uses_additional_data":false,"paper_date":"2020-03-18","paper":"/paper/stem-seg-spatio-temporal-embeddings-for","paper_url":"https://arxiv.org/abs/2003.08429v4","paper_title":"STEm-Seg: Spatio-temporal Embeddings for Instance Segmentation in Videos","code":"https://github.com/sabarim/STEm-Seg","n_code_links":1,"syntology":null},{"rank_in_archive_order":41,"model":"MaskTrack R-CNN (ResNet-50, single-scale training and test)","metrics":{"AP50":"51.1","AP75":"32.6","AR1":"31","AR10":"35.5","mask AP":"30.3"},"uses_additional_data":false,"paper_date":"2019-05-12","paper":"/paper/video-instance-segmentation","paper_url":"https://arxiv.org/abs/1905.04804v4","paper_title":"Video Instance Segmentation","code":"https://github.com/JonathonLuiten/HOTA-metrics","n_code_links":6,"syntology":{"n_ran":5,"n_unverified":6,"n_samples":11,"n_pointer_only_licence":1}},{"rank_in_archive_order":42,"model":"UniTrack","metrics":{"mask AP":"30.1"},"uses_additional_data":false,"paper_date":"2021-07-05","paper":"/paper/do-different-tracking-tasks-require-different","paper_url":"https://arxiv.org/abs/2107.02156v2","paper_title":"Do Different Tracking Tasks Require Different Appearance Models?","code":"https://github.com/Zhongdao/UniTrack","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":11,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":43,"model":"OSMN","metrics":{"AP50":"28.6","AP75":"33.1","mask AP":"29.1"},"uses_additional_data":false,"paper_date":"2018-02-04","paper":"/paper/efficient-video-object-segmentation-via","paper_url":"http://arxiv.org/abs/1802.01218v1","paper_title":"Efficient Video Object Segmentation via Network Modulation","code":"https://github.com/linjieyangsc/video_seg","n_code_links":1,"syntology":null},{"rank_in_archive_order":44,"model":"DeepSORT","metrics":{"AP50":"31.3","mask AP":"27.8"},"uses_additional_data":false,"paper_date":"2017-03-21","paper":"/paper/simple-online-and-realtime-tracking-with-a","paper_url":"http://arxiv.org/abs/1703.07402v1","paper_title":"Simple Online and Realtime Tracking with a Deep Association Metric","code":"https://github.com/PaddlePaddle/PaddleDetection","n_code_links":75,"syntology":{"n_ran":15,"n_unverified":25,"n_samples":40,"n_pointer_only_licence":7}}],"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 6,885 of the 9,623 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":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+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":14,"rows_with_any_sample_ran":13,"distinct_papers_with_graph_line":11,"distinct_papers_with_any_sample_ran":10,"samples_over_distinct_papers":{"n_ran":62,"n_unverified":59,"n_samples":121,"n_pointer_only_licence":37,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":75,"n_unverified":71,"n_samples":146,"n_pointer_only_licence":48,"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"}}}