{"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/murauer-mapping-unlabeled-real-data-for-label","title":"MURAUER: Mapping Unlabeled Real Data for Label AUstERity","arxiv_id":"1811.09497","date":"2018-11-23","proceeding":null,"authors":["Georg Poier","Michael Opitz","David Schinagl","Horst Bischof"],"abstract":"Data labeling for learning 3D hand pose estimation models is a huge effort.\nReadily available, accurately labeled synthetic data has the potential to\nreduce the effort. However, to successfully exploit synthetic data, current\nstate-of-the-art methods still require a large amount of labeled real data. In\nthis work, we remove this requirement by learning to map from the features of\nreal data to the features of synthetic data mainly using a large amount of\nsynthetic and unlabeled real data. We exploit unlabeled data using two\nauxiliary objectives, which enforce that (i) the mapped representation is pose\nspecific and (ii) at the same time, the distributions of real and synthetic\ndata are aligned. While pose specifity is enforced by a self-supervisory signal\nrequiring that the representation is predictive for the appearance from\ndifferent views, distributions are aligned by an adversarial term. In this way,\nwe can significantly improve the results of the baseline system, which does not\nuse unlabeled data and outperform many recent approaches already with about 1%\nof the labeled real data. This presents a step towards faster deployment of\nlearning based hand pose estimation, making it accessible for a larger range of\napplications.","url_abs":"http://arxiv.org/abs/1811.09497v2","url_pdf":"http://arxiv.org/pdf/1811.09497v2.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":"murauer-mapping-unlabeled-real-data-for-label","repo_url":"https://github.com/poier/murauer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-hand-pose-estimation","task_name":"3D Hand Pose Estimation"},{"task_slug":"hand-pose-estimation","task_name":"Hand Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}