{"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-joint-reconstruction-of-hands-and","title":"Learning joint reconstruction of hands and manipulated objects","arxiv_id":"1904.05767","date":"2019-04-11","proceeding":"CVPR 2019 6","authors":["Yana Hasson","Gül Varol","Dimitrios Tzionas","Igor Kalevatykh","Michael J. Black","Ivan Laptev","Cordelia Schmid"],"abstract":"Estimating hand-object manipulations is essential for interpreting and\nimitating human actions. Previous work has made significant progress towards\nreconstruction of hand poses and object shapes in isolation. Yet,\nreconstructing hands and objects during manipulation is a more challenging task\ndue to significant occlusions of both the hand and object. While presenting\nchallenges, manipulations may also simplify the problem since the physics of\ncontact restricts the space of valid hand-object configurations. For example,\nduring manipulation, the hand and object should be in contact but not\ninterpenetrate. In this work, we regularize the joint reconstruction of hands\nand objects with manipulation constraints. We present an end-to-end learnable\nmodel that exploits a novel contact loss that favors physically plausible\nhand-object constellations. Our approach improves grasp quality metrics over\nbaselines, using RGB images as input. To train and evaluate the model, we also\npropose a new large-scale synthetic dataset, ObMan, with hand-object\nmanipulations. We demonstrate the transferability of ObMan-trained models to\nreal data.","url_abs":"http://arxiv.org/abs/1904.05767v1","url_pdf":"http://arxiv.org/pdf/1904.05767v1.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-joint-reconstruction-of-hands-and","repo_url":"https://github.com/hassony2/manopth","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"learning-joint-reconstruction-of-hands-and","repo_url":"https://github.com/hassony2/obman","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"learning-joint-reconstruction-of-hands-and","repo_url":"https://github.com/hassony2/obman_train","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"3d-hand-pose-estimation","task_name":"3D Hand Pose Estimation"},{"task_slug":"hand-joint-reconstruction","task_name":"Hand Joint Reconstruction"},{"task_slug":"object","task_name":"Object"},{"task_slug":"hand-object-pose","task_name":"hand-object pose"},{"task_slug":null,"task_name":"valid"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-hand-pose-estimation-on-freihand","task":"3D Hand Pose Estimation","dataset":"FreiHAND","model":"Hasson et al.","rank_in_archive_order":33,"of":33,"metrics":{"PA-F@15mm":"0.908","PA-F@5mm":"0.436","PA-MPVPE":"13.2"},"uses_additional_data":false},{"leaderboard":"/sota/hand-object-pose-on-dexycb","task":"hand-object pose","dataset":"DexYCB","model":"HMO","rank_in_archive_order":7,"of":9,"metrics":{"ADD-S":"-","Average MPJPE (mm)":"17.6","MCE":"-","OCE":"-","Procrustes-Aligned MPJPE":"-"},"uses_additional_data":false},{"leaderboard":"/sota/hand-object-pose-on-ho-3d","task":"hand-object pose","dataset":"HO-3D v2","model":"HMO","rank_in_archive_order":8,"of":9,"metrics":{"ADD-S":"-","Average MPJPE (mm)":"-","OME":"-","PA-MPJPE":"11.0","ST-MPJPE":"31.8"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1904.05767","atlas_url":"https://app.syntology.ai/?focus=1904.05767","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}