{"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/learning-correspondence-from-the-cycle","title":"Learning Correspondence from the Cycle-Consistency of Time","arxiv_id":"1903.07593","date":"2019-03-18","proceeding":"CVPR 2019 6","authors":["Xiaolong Wang","Allan Jabri","Alexei A. Efros"],"abstract":"We introduce a self-supervised method for learning visual correspondence from\nunlabeled video. The main idea is to use cycle-consistency in time as free\nsupervisory signal for learning visual representations from scratch. At\ntraining time, our model learns a feature map representation to be useful for\nperforming cycle-consistent tracking. At test time, we use the acquired\nrepresentation to find nearest neighbors across space and time. We demonstrate\nthe generalizability of the representation -- without finetuning -- across a\nrange of visual correspondence tasks, including video object segmentation,\nkeypoint tracking, and optical flow. Our approach outperforms previous\nself-supervised methods and performs competitively with strongly supervised\nmethods.","url_abs":"http://arxiv.org/abs/1903.07593v2","url_pdf":"http://arxiv.org/pdf/1903.07593v2.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":"learning-correspondence-from-the-cycle","repo_url":"https://github.com/xiaolonw/TimeCycle","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"semi-supervised-video-object-segmentation","task_name":"Semi-Supervised Video Object Segmentation"},{"task_slug":"unsupervised-video-object-segmentation","task_name":"Unsupervised 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-2017","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (val)","model":"CycleTime","rank_in_archive_order":80,"of":81,"metrics":{"F-measure (Mean)":"50.0","F-measure (Recall)":"48.0","J&F":"48.7","Jaccard (Mean)":"46.4","Jaccard (Recall)":"50.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.07593","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}