{"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-human-optical-flow","title":"Learning Human Optical Flow","arxiv_id":"1806.05666","date":"2018-06-14","proceeding":null,"authors":["Anurag Ranjan","Javier Romero","Michael J. Black"],"abstract":"The optical flow of humans is well known to be useful for the analysis of\nhuman action. Given this, we devise an optical flow algorithm specifically for\nhuman motion and show that it is superior to generic flow methods. Designing a\nmethod by hand is impractical, so we develop a new training database of image\nsequences with ground truth optical flow. For this we use a 3D model of the\nhuman body and motion capture data to synthesize realistic flow fields. We then\ntrain a convolutional neural network to estimate human flow fields from pairs\nof images. Since many applications in human motion analysis depend on speed,\nand we anticipate mobile applications, we base our method on SpyNet with\nseveral modifications. We demonstrate that our trained network is more accurate\nthan a wide range of top methods on held-out test data and that it generalizes\nwell to real image sequences. When combined with a person detector/tracker, the\napproach provides a full solution to the problem of 2D human flow estimation.\nBoth the code and the dataset are available for research.","url_abs":"http://arxiv.org/abs/1806.05666v2","url_pdf":"http://arxiv.org/pdf/1806.05666v2.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-human-optical-flow","repo_url":"https://github.com/anuragranj/humanflow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"}],"methods":[],"datasets_introduced":[{"slug":"coma-1","name":"Human Optical Flow","full_name":"Human Optical Flow dataset"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.05666","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}