{"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/seqformer-a-frustratingly-simple-model-for","title":"SeqFormer: Sequential Transformer for Video Instance Segmentation","arxiv_id":"2112.08275","date":"2021-12-15","proceeding":null,"authors":["Junfeng Wu","Yi Jiang","Song Bai","Wenqing Zhang","Xiang Bai"],"abstract":"In this work, we present SeqFormer for video instance segmentation. SeqFormer follows the principle of vision transformer that models instance relationships among video frames. Nevertheless, we observe that a stand-alone instance query suffices for capturing a time sequence of instances in a video, but attention mechanisms shall be done with each frame independently. To achieve this, SeqFormer locates an instance in each frame and aggregates temporal information to learn a powerful representation of a video-level instance, which is used to predict the mask sequences on each frame dynamically. Instance tracking is achieved naturally without tracking branches or post-processing. On YouTube-VIS, SeqFormer achieves 47.4 AP with a ResNet-50 backbone and 49.0 AP with a ResNet-101 backbone without bells and whistles. Such achievement significantly exceeds the previous state-of-the-art performance by 4.6 and 4.4, respectively. In addition, integrated with the recently-proposed Swin transformer, SeqFormer achieves a much higher AP of 59.3. We hope SeqFormer could be a strong baseline that fosters future research in video instance segmentation, and in the meantime, advances this field with a more robust, accurate, neat model. The code is available at https://github.com/wjf5203/SeqFormer.","url_abs":"https://arxiv.org/abs/2112.08275v2","url_pdf":"https://arxiv.org/pdf/2112.08275v2.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":"seqformer-a-frustratingly-simple-model-for","repo_url":"https://github.com/wjf5203/SeqFormer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"seqformer-a-frustratingly-simple-model-for","repo_url":"https://github.com/wjf5203/vnext","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"video-instance-segmentation","task_name":"Video Instance Segmentation"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-instance-segmentation-on-hq-ytvis","task":"Video Instance Segmentation","dataset":"HQ-YTVIS","model":"SeqFormer (Swin-L)","rank_in_archive_order":2,"of":4,"metrics":{"Tube-Boundary AP":"43.3"},"uses_additional_data":false},{"leaderboard":"/sota/video-instance-segmentation-on-youtube-vis-1","task":"Video Instance Segmentation","dataset":"YouTube-VIS validation","model":"SeqFormer (Swin-L)","rank_in_archive_order":9,"of":44,"metrics":{"AP50":"82.1","AP75":"66.4","AR1":"51.7","AR10":"64.4","mask AP":"59.3"},"uses_additional_data":true},{"leaderboard":"/sota/video-instance-segmentation-on-youtube-vis-1","task":"Video Instance Segmentation","dataset":"YouTube-VIS validation","model":"SeqFormer (ResNet-101)","rank_in_archive_order":17,"of":44,"metrics":{"AP50":"71.1","AP75":"55.7","AR1":"46.8","AR10":"56.9","mask AP":"49.0"},"uses_additional_data":true},{"leaderboard":"/sota/video-instance-segmentation-on-youtube-vis-1","task":"Video Instance Segmentation","dataset":"YouTube-VIS validation","model":"SeqFormer (ResNet-50)","rank_in_archive_order":19,"of":44,"metrics":{"AP50":"69.8","AP75":"51.8","AR1":"45.5","AR10":"54.8","mask AP":"47.4"},"uses_additional_data":true},{"leaderboard":"/sota/video-instance-segmentation-on-youtube-vis-1","task":"Video Instance Segmentation","dataset":"YouTube-VIS validation","model":"SeqFormer (ResNet-50)","rank_in_archive_order":22,"of":44,"metrics":{"AP50":"66.9","AP75":"50.5","AR1":"45.6","AR10":"54.6","mask AP":"45.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2112.08275","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}