{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,14]],"date-time":"2025-10-14T07:17:37Z","timestamp":1760426257724,"version":"3.28.0"},"reference-count":38,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,5,13]],"date-time":"2024-05-13T00:00:00Z","timestamp":1715558400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,5,13]],"date-time":"2024-05-13T00:00:00Z","timestamp":1715558400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,5,13]]},"DOI":"10.1109\/icra57147.2024.10611423","type":"proceedings-article","created":{"date-parts":[[2024,8,8]],"date-time":"2024-08-08T17:51:05Z","timestamp":1723139465000},"page":"5375-5381","source":"Crossref","is-referenced-by-count":5,"title":["RoboKeyGen: Robot Pose and Joint Angles Estimation via Diffusion-based 3D Keypoint Generation"],"prefix":"10.1109","author":[{"given":"Yang","family":"Tian","sequence":"first","affiliation":[{"name":"Peking University and National Key Laboratory for Multimedia Information Processing,CFCS, School of CS"}]},{"given":"Jiyao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Peking University and National Key Laboratory for Multimedia Information Processing,CFCS, School of CS"}]},{"given":"Guowei","family":"Huang","sequence":"additional","affiliation":[{"name":"Huawei"}]},{"given":"Bin","family":"Wang","sequence":"additional","affiliation":[{"name":"Huawei"}]},{"given":"Ping","family":"Wang","sequence":"additional","affiliation":[{"name":"Peking University,School of Software & Microelectronics and National Engineering Research Center for Software Engineering"}]},{"given":"Jiangmiao","family":"Pang","sequence":"additional","affiliation":[{"name":"Chinese University of Hong Kong"}]},{"given":"Hao","family":"Dong","sequence":"additional","affiliation":[{"name":"Peking University and National Key Laboratory for Multimedia Information Processing,CFCS, School of CS"}]}],"member":"263","reference":[{"key":"ref1","first-page":"1","article-title":"Cooperative heterogeneous multi-robot systems","volume-title":"ACM Computing Surveys (CSUR)","volume":"52","author":"Rizk","year":"2019"},{"key":"ref2","article-title":"Event-driven visual-tactile sensing and learning for robots","volume-title":"ArXiv","volume":"abs\/2009.07083","author":"Taunyazov","year":"2020"},{"article-title":"Image moments : a general and useful set of features for visual servoing","year":"2017","author":"Chaumette","key":"ref3"},{"key":"ref4","first-page":"345","article-title":"A new technique for fully autonomous and efficient 3d robotics hand\/eye calibration","volume-title":"IEEE Trans. Robotics Autom","volume":"5","author":"Tsai","year":"1988"},{"key":"ref5","first-page":"9426","article-title":"Camera-to-robot pose estimation from a single image","volume-title":"2020 IEEE International Conference on Robotics and Automation (ICRA)","author":"Lee"},{"key":"ref6","first-page":"8917","article-title":"Robot structure prior guided temporal attention for camera-to-robot pose estimation from image sequence","volume-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Tian"},{"key":"ref7","first-page":"21 296","article-title":"Markerless camera-to-robot pose estimation via self-supervised sim-to-real transfer","volume-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Lu"},{"key":"ref8","first-page":"1654","article-title":"Single-view robot pose and joint angle estimation via render & compare","volume-title":"2021 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Labb\u2019e"},{"key":"ref9","first-page":"11 569","article-title":"Semi-perspective decoupled heatmaps for 3d robot pose estimation from depth maps","volume-title":"IEEE Robotics and Automation Letters","volume":"7","author":"Simoni","year":"2022"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01231-1_42"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00203"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00275"},{"article-title":"Genpose: Generative category-level object pose estimation via diffusion models","year":"2023","author":"Zhang","key":"ref13"},{"key":"ref14","first-page":"2886","article-title":"Rede: End-to-end object 6d pose robust estimation using differentiable outliers elimination","volume-title":"IEEE Robotics and Automation Letters","volume":"6","author":"Hua","year":"2020"},{"key":"ref15","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-58580-8_22","article-title":"Learning gradient fields for shape generation","volume-title":"European Conference on Computer Vision","author":"Cai"},{"key":"ref16","article-title":"Targf: Learning target gradient field for object rearrangement","volume-title":"ArXiv","volume":"abs\/2209.00853","author":"Wu","year":"2022"},{"key":"ref17","article-title":"A review of solutions for perspective-n-point problem in camera pose estimation","volume-title":"Journal of Physics: Conference Series","volume":"1087","author":"Lu"},{"key":"ref18","article-title":"Generative modeling by estimating gradients of the data distribution","volume-title":"Neural Information Processing Systems","author":"Song","year":"2019"},{"key":"ref19","first-page":"1661","article-title":"A connection between score matching and denoising autoencoders","volume-title":"Neural Computation","volume":"23","author":"Vincent","year":"2011"},{"key":"ref20","article-title":"Denoising diffusion probabilistic models","volume-title":"ArXiv","volume":"abs\/2006.11239","author":"Ho","year":"2020"},{"key":"ref21","article-title":"Deep unsupervised learning using nonequilibrium thermodynamics","volume-title":"ArXiv","volume":"abs\/1503.03585","author":"Sohl-Dickstein","year":"2015"},{"key":"ref22","article-title":"Score-based generative modeling through stochastic differential equations","volume-title":"ArXiv","volume":"abs\/2011.13456","author":"Song","year":"2020"},{"key":"ref23","article-title":"Solving inverse problems in medical imaging with score-based generative models","volume-title":"ArXiv","volume":"abs\/2111.08005","author":"Song","year":"2021"},{"article-title":"Example-based planning via dual gradient fields","year":"2022","author":"Wu","key":"ref24"},{"key":"ref25","article-title":"Genpose: Generative category-level object pose estimation via diffusion models","volume-title":"ArXiv","volume":"abs\/2306.10531","author":"Zhang","year":"2023"},{"key":"ref26","first-page":"4800","article-title":"Gfpose: Learning 3d human pose prior with gradient fields","volume-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Ci"},{"key":"ref27","article-title":"Pidnet: A real-time semantic segmentation network inspired from pid controller","volume-title":"ArXiv","volume":"abs\/2206.02066","author":"Xu","year":"2022"},{"key":"ref28","article-title":"Attention is all you need","volume-title":"NIPS","author":"Vaswani","year":"2017"},{"key":"ref29","article-title":"Rtmpose: Real-time multi-person pose estimation based on mmpose","volume-title":"ArXiv","volume":"abs\/2303.07399","author":"Jiang","year":"2023"},{"key":"ref30","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-20068-7_6","article-title":"Simcc: A simple coordinate classification perspective for human pose estimation","volume-title":"European Conference on Computer Vision","author":"Li"},{"key":"ref31","article-title":"Maximum likelihood training of score-based diffusion models","volume-title":"Neural Information Processing Systems","author":"Song","year":"2021"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref33","article-title":"Denoising diffusion implicit models","volume-title":"ArXiv","volume":"abs\/2010.02502","author":"Song","year":"2020"},{"key":"ref34","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-19842-7_22","article-title":"Domain randomization-enhanced depth simulation and restoration for perceiving and grasping specular and transparent objects","volume-title":"European Conference on Computer Vision","author":"Dai"},{"key":"ref35","article-title":"Blender"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.445"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.288"},{"key":"ref38","first-page":"19","article-title":"A family of embedded runge-kutta formulae","volume-title":"Journal of Computational and Applied Mathematics","volume":"6","author":"Dormand","year":"1980"}],"event":{"name":"2024 IEEE International Conference on Robotics and Automation (ICRA)","start":{"date-parts":[[2024,5,13]]},"location":"Yokohama, Japan","end":{"date-parts":[[2024,5,17]]}},"container-title":["2024 IEEE International Conference on Robotics and Automation (ICRA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10609961\/10609862\/10611423.pdf?arnumber=10611423","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,11]],"date-time":"2024-08-11T04:17:46Z","timestamp":1723349866000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10611423\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,5,13]]},"references-count":38,"URL":"https:\/\/doi.org\/10.1109\/icra57147.2024.10611423","relation":{},"subject":[],"published":{"date-parts":[[2024,5,13]]}}}