{"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/3d-hand-shape-and-pose-estimation-from-a","title":"3D Hand Shape and Pose Estimation from a Single RGB Image","arxiv_id":"1903.00812","date":"2019-03-03","proceeding":"CVPR 2019 6","authors":["Liuhao Ge","Zhou Ren","Yuncheng Li","Zehao Xue","Yingying Wang","Jianfei Cai","Junsong Yuan"],"abstract":"This work addresses a novel and challenging problem of estimating the full 3D\nhand shape and pose from a single RGB image. Most current methods in 3D hand\nanalysis from monocular RGB images only focus on estimating the 3D locations of\nhand keypoints, which cannot fully express the 3D shape of hand. In contrast,\nwe propose a Graph Convolutional Neural Network (Graph CNN) based method to\nreconstruct a full 3D mesh of hand surface that contains richer information of\nboth 3D hand shape and pose. To train networks with full supervision, we create\na large-scale synthetic dataset containing both ground truth 3D meshes and 3D\nposes. When fine-tuning the networks on real-world datasets without 3D ground\ntruth, we propose a weakly-supervised approach by leveraging the depth map as a\nweak supervision in training. Through extensive evaluations on our proposed new\ndatasets and two public datasets, we show that our proposed method can produce\naccurate and reasonable 3D hand mesh, and can achieve superior 3D hand pose\nestimation accuracy when compared with state-of-the-art methods.","url_abs":"http://arxiv.org/abs/1903.00812v2","url_pdf":"http://arxiv.org/pdf/1903.00812v2.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":"3d-hand-shape-and-pose-estimation-from-a","repo_url":"https://github.com/3d-hand-shape/hand-graph-cnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"3d-hand-shape-and-pose-estimation-from-a","repo_url":"https://github.com/ArashHosseini/hand-graph-cnn-gpu","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"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":"https://app.syntology.ai/?focus=1903.00812","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}