{"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/continual-occlusions-and-optical-flow","title":"Continual Occlusions and Optical Flow Estimation","arxiv_id":"1811.01602","date":"2018-11-05","proceeding":null,"authors":["Michal Neoral","Jan Šochman","Jiří Matas"],"abstract":"Two optical flow estimation problems are addressed: i) occlusion estimation\nand handling, and ii) estimation from image sequences longer than two frames.\nThe proposed ContinualFlow method estimates occlusions before flow, avoiding\nthe use of flow corrupted by occlusions for their estimation. We show that\nproviding occlusion masks as an additional input to flow estimation improves\nthe standard performance metric by more than 25\\% on both KITTI and Sintel. As\na second contribution, a novel method for incorporating information from past\nframes into flow estimation is introduced. The previous frame flow serves as an\ninput to occlusion estimation and as a prior in occluded regions, i.e. those\nwithout visual correspondences. By continually using the previous frame flow,\nContinualFlow performance improves further by 18\\% on KITTI and 7\\% on Sintel,\nachieving top performance on KITTI and Sintel.","url_abs":"http://arxiv.org/abs/1811.01602v1","url_pdf":"http://arxiv.org/pdf/1811.01602v1.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":[],"tasks":[{"task_slug":"occlusion-estimation","task_name":"Occlusion Estimation"},{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/optical-flow-estimation-on-sintel-final","task":"Optical Flow Estimation","dataset":"Sintel-final","model":"ContinualFlow + ft","rank_in_archive_order":19,"of":28,"metrics":{"Average End-Point Error":"4.52"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.01602","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}