{"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/unsupervised-learning-of-object-landmarks-by","title":"Unsupervised learning of object landmarks by factorized spatial embeddings","arxiv_id":"1705.02193","date":"2017-05-05","proceeding":"ICCV 2017 10","authors":["James Thewlis","Hakan Bilen","Andrea Vedaldi"],"abstract":"Learning automatically the structure of object categories remains an\nimportant open problem in computer vision. In this paper, we propose a novel\nunsupervised approach that can discover and learn landmarks in object\ncategories, thus characterizing their structure. Our approach is based on\nfactorizing image deformations, as induced by a viewpoint change or an object\ndeformation, by learning a deep neural network that detects landmarks\nconsistently with such visual effects. Furthermore, we show that the learned\nlandmarks establish meaningful correspondences between different object\ninstances in a category without having to impose this requirement explicitly.\nWe assess the method qualitatively on a variety of object types, natural and\nman-made. We also show that our unsupervised landmarks are highly predictive of\nmanually-annotated landmarks in face benchmark datasets, and can be used to\nregress these with a high degree of accuracy.","url_abs":"http://arxiv.org/abs/1705.02193v2","url_pdf":"http://arxiv.org/pdf/1705.02193v2.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":"unsupervised-learning-of-object-landmarks-by","repo_url":"https://github.com/alldbi/Factorized-Spatial-Embeddings","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"unsupervised-facial-landmark-detection","task_name":"Unsupervised Facial Landmark Detection"},{"task_slug":"unsupervised-human-pose-estimation","task_name":"Unsupervised Human Pose Estimation"},{"task_slug":null,"task_name":"Unsupervised Keypoints"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/unsupervised-facial-landmark-detection-on","task":"Unsupervised Facial Landmark Detection","dataset":"300W","model":"FSE","rank_in_archive_order":3,"of":4,"metrics":{"NME":"7.97"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-facial-landmark-detection-on-3","task":"Unsupervised Facial Landmark Detection","dataset":"AFLW-MTFL","model":"FSE","rank_in_archive_order":2,"of":3,"metrics":{"NME":"10.53"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-facial-landmark-detection-on-1","task":"Unsupervised Facial Landmark Detection","dataset":"MAFL","model":"Thewlis2017unsupervised","rank_in_archive_order":11,"of":13,"metrics":{"NME":"6.32"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-facial-landmark-detection-on-1","task":"Unsupervised Facial Landmark Detection","dataset":"MAFL","model":"FSE","rank_in_archive_order":12,"of":13,"metrics":{"NME":"6.67"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-facial-landmark-detection-on-5","task":"Unsupervised Facial Landmark Detection","dataset":"MAFL Unaligned","model":"ULD","rank_in_archive_order":8,"of":9,"metrics":{"NME":"31.3"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1705.02193","atlas_url":"https://app.syntology.ai/?focus=1705.02193","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}