{"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/social-fabric-tubelet-compositions-for-video","title":"Social Fabric: Tubelet Compositions for Video Relation Detection","arxiv_id":"2108.08363","date":"2021-08-18","proceeding":"ICCV 2021 10","authors":["Shuo Chen","Zenglin Shi","Pascal Mettes","Cees G. M. Snoek"],"abstract":"This paper strives to classify and detect the relationship between object tubelets appearing within a video as a <subject-predicate-object> triplet. Where existing works treat object proposals or tubelets as single entities and model their relations a posteriori, we propose to classify and detect predicates for pairs of object tubelets a priori. We also propose Social Fabric: an encoding that represents a pair of object tubelets as a composition of interaction primitives. These primitives are learned over all relations, resulting in a compact representation able to localize and classify relations from the pool of co-occurring object tubelets across all timespans in a video. The encoding enables our two-stage network. In the first stage, we train Social Fabric to suggest proposals that are likely interacting. We use the Social Fabric in the second stage to simultaneously fine-tune and predict predicate labels for the tubelets. Experiments demonstrate the benefit of early video relation modeling, our encoding and the two-stage architecture, leading to a new state-of-the-art on two benchmarks. We also show how the encoding enables query-by-primitive-example to search for spatio-temporal video relations. Code: https://github.com/shanshuo/Social-Fabric.","url_abs":"https://arxiv.org/abs/2108.08363v1","url_pdf":"https://arxiv.org/pdf/2108.08363v1.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":"social-fabric-tubelet-compositions-for-video","repo_url":"https://github.com/shanshuo/social-fabric","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":null,"task_name":"Relation"},{"task_slug":null,"task_name":"Triplet"},{"task_slug":"video-visual-relation-detection","task_name":"Video Visual Relation Detection"},{"task_slug":"video-visual-relation-tagging","task_name":"Video Visual Relation Tagging"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-visual-relation-detection-on-imagenet","task":"Video Visual Relation Detection","dataset":"ImageNet-VidVRD","model":"Social Fabric","rank_in_archive_order":1,"of":2,"metrics":{"Recall@100":"16.88","Recall@50":"13.73","mAP":"20.08"},"uses_additional_data":false},{"leaderboard":"/sota/video-visual-relation-detection-on-vidor","task":"Video Visual Relation Detection","dataset":"VidOR","model":"Social Fabric","rank_in_archive_order":1,"of":2,"metrics":{"Recall@100":"11.94","Recall@50":"9.99","mAP":"11.21"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2108.08363","atlas_url":"https://app.syntology.ai/?focus=2108.08363","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}