{"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/rethinking-space-time-networks-with-improved","title":"Rethinking Space-Time Networks with Improved Memory Coverage for Efficient Video Object Segmentation","arxiv_id":"2106.05210","date":"2021-06-09","proceeding":"NeurIPS 2021 12","authors":["Ho Kei Cheng","Yu-Wing Tai","Chi-Keung Tang"],"abstract":"This paper presents a simple yet effective approach to modeling space-time correspondences in the context of video object segmentation. Unlike most existing approaches, we establish correspondences directly between frames without re-encoding the mask features for every object, leading to a highly efficient and robust framework. With the correspondences, every node in the current query frame is inferred by aggregating features from the past in an associative fashion. We cast the aggregation process as a voting problem and find that the existing inner-product affinity leads to poor use of memory with a small (fixed) subset of memory nodes dominating the votes, regardless of the query. In light of this phenomenon, we propose using the negative squared Euclidean distance instead to compute the affinities. We validated that every memory node now has a chance to contribute, and experimentally showed that such diversified voting is beneficial to both memory efficiency and inference accuracy. The synergy of correspondence networks and diversified voting works exceedingly well, achieves new state-of-the-art results on both DAVIS and YouTubeVOS datasets while running significantly faster at 20+ FPS for multiple objects without bells and whistles.","url_abs":"https://arxiv.org/abs/2106.05210v2","url_pdf":"https://arxiv.org/pdf/2106.05210v2.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":"rethinking-space-time-networks-with-improved","repo_url":"https://github.com/hkchengrex/STCN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"rethinking-space-time-networks-with-improved","repo_url":"https://github.com/alipga/AMM_VOS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"rethinking-space-time-networks-with-improved","repo_url":"https://github.com/limingxing00/rde-vos-cvpr2022","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"semi-supervised-video-object-segmentation","task_name":"Semi-Supervised Video Object Segmentation"},{"task_slug":"video-object-segmentation","task_name":"Video Object Segmentation"},{"task_slug":"video-semantic-segmentation","task_name":"Video Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/visual-object-tracking-on-davis-2016","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2016","model":"STCN","rank_in_archive_order":18,"of":78,"metrics":{"F-measure (Decay)":"4.3","F-measure (Mean)":"93.0","F-measure (Recall)":"97.1","J&F":"91.7","Jaccard (Decay)":"4.1","Jaccard (Mean)":"90.4","Jaccard (Recall)":"98.1","Speed (FPS)":"26.9"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-1","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (test-dev)","model":"STCN","rank_in_archive_order":19,"of":59,"metrics":{"F-measure (Decay)":"10.3","F-measure (Mean)":"83.5","F-measure (Recall)":"89.7","J&F":"79.9","Jaccard (Decay)":"10.5","Jaccard (Mean)":"76.3","Jaccard (Recall)":"85.5"},"uses_additional_data":true},{"leaderboard":"/sota/visual-object-tracking-on-davis-2017","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (val)","model":"STCN","rank_in_archive_order":23,"of":81,"metrics":{"F-measure (Decay)":"85.3","F-measure (Mean)":"88.6","F-measure (Recall)":"94.6","J&F":"85.3","Jaccard (Decay)":"6.2","Jaccard (Mean)":"82.0","Jaccard (Recall)":"91.3","Speed (FPS)":"20.2"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"STCN","rank_in_archive_order":16,"of":17,"metrics":{"F":"55.0","J":"46.6","J&F":"50.8"},"uses_additional_data":false},{"leaderboard":"/sota/video-object-segmentation-on-youtube-vos","task":"Semi-Supervised Video Object Segmentation","dataset":"YouTube-VOS 2018","model":"STCN","rank_in_archive_order":53,"of":53,"metrics":{"F-Measure (Seen)":"87.9","F-Measure (Unseen)":"87.3","Jaccard (Seen)":"83.2","Jaccard (Unseen)":"79.0"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-18","task":"Semi-Supervised Video Object Segmentation","dataset":"YouTube-VOS 2019","model":"STCN (MS)","rank_in_archive_order":10,"of":22,"metrics":{"F-Measure (Seen)":"87.8","F-Measure (Unseen)":"88.8","Jaccard (Seen)":"83.5","Jaccard (Unseen)":"80.8","Overall":"85.2"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-18","task":"Semi-Supervised Video Object Segmentation","dataset":"YouTube-VOS 2019","model":"STCN","rank_in_archive_order":16,"of":22,"metrics":{"F-Measure (Seen)":"87.0","F-Measure (Unseen)":"87.7","Jaccard (Seen)":"82.6","Jaccard (Unseen)":"79.4","Overall":"84.2"},"uses_additional_data":false},{"leaderboard":"/sota/video-object-segmentation-on-youtube-vos-2019-2","task":"Video Object Segmentation","dataset":"YouTube-VOS 2019","model":"STCN","rank_in_archive_order":7,"of":10,"metrics":{"F-Measure (Seen)":"85.4","F-Measure (Unseen)":"85.9","Jaccard (Seen)":"81.1","Jaccard (Unseen)":"78.2","Mean Jaccard & F-Measure":"82.7"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2106.05210","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.05210"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/alipga/AMM_VOS","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/limingxing00/rde-vos-cvpr2022","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hkchengrex/STCN","reach":null}],"summary":{"ran":6,"unverified":4},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1},"listed":{"samples":9,"ran":6,"repositories":2}},"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":2,"samples":[{"code_sha256_prefix":"812c10c240586d4c","entry":"ASPP3D","repo":"limingxing00/rde-vos-cvpr2022","repo_kind":"listed","path":"model/modules.py","file_url":"https://github.com/limingxing00/rde-vos-cvpr2022/blob/HEAD/model/modules.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"812c10c240586d4c"}},{"code_sha256_prefix":"c5daadfa598c8697","entry":"NONLocalBlock3D","repo":"limingxing00/rde-vos-cvpr2022","repo_kind":"listed","path":"model/modules.py","file_url":"https://github.com/limingxing00/rde-vos-cvpr2022/blob/HEAD/model/modules.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c5daadfa598c8697"}},{"code_sha256_prefix":"01ec14b944f4999c","entry":"SEBasicBlock","repo":"limingxing00/rde-vos-cvpr2022","repo_kind":"listed","path":"model/modules.py","file_url":"https://github.com/limingxing00/rde-vos-cvpr2022/blob/HEAD/model/modules.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"01ec14b944f4999c"}},{"code_sha256_prefix":"9bae3aa52ec50d1a","entry":"SELayerS","repo":"limingxing00/rde-vos-cvpr2022","repo_kind":"listed","path":"model/modules.py","file_url":"https://github.com/limingxing00/rde-vos-cvpr2022/blob/HEAD/model/modules.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9bae3aa52ec50d1a"}},{"code_sha256_prefix":"698c6d37906b6b77","entry":"_ASPPModule3D","repo":"limingxing00/rde-vos-cvpr2022","repo_kind":"listed","path":"model/modules.py","file_url":"https://github.com/limingxing00/rde-vos-cvpr2022/blob/HEAD/model/modules.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"698c6d37906b6b77"}},{"code_sha256_prefix":"42805181cd60008e","entry":"_NonLocalBlockND","repo":"limingxing00/rde-vos-cvpr2022","repo_kind":"listed","path":"model/modules.py","file_url":"https://github.com/limingxing00/rde-vos-cvpr2022/blob/HEAD/model/modules.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"42805181cd60008e"}},{"code_sha256_prefix":"6618c26355cb868a","entry":"MemoryReader","repo":"hkchengrex/STCN","repo_kind":"official","path":"model/network.py","file_url":"https://github.com/hkchengrex/STCN/blob/HEAD/model/network.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6618c26355cb868a"}},{"code_sha256_prefix":"fa9203b36ccb7cee","entry":"MemoryReader","repo":"alipga/AMM_VOS","repo_kind":"listed","path":"model/network.py","file_url":"https://github.com/alipga/AMM_VOS/blob/HEAD/model/network.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fa9203b36ccb7cee"}},{"code_sha256_prefix":"aed6d31ae6227bce","entry":"MyDropout","repo":"alipga/AMM_VOS","repo_kind":"listed","path":"model/network.py","file_url":"https://github.com/alipga/AMM_VOS/blob/HEAD/model/network.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"aed6d31ae6227bce"}},{"code_sha256_prefix":"2a010015f43898d9","entry":"SAM","repo":"limingxing00/rde-vos-cvpr2022","repo_kind":"listed","path":"model/modules.py","file_url":"https://github.com/limingxing00/rde-vos-cvpr2022/blob/HEAD/model/modules.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2a010015f43898d9"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}