{"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-3d-dynamic-scene-representations-for","title":"Learning 3D Dynamic Scene Representations for Robot Manipulation","arxiv_id":"2011.01968","date":"2020-11-03","proceeding":null,"authors":["Zhenjia Xu","Zhanpeng He","Jiajun Wu","Shuran Song"],"abstract":"3D scene representation for robot manipulation should capture three key object properties: permanency -- objects that become occluded over time continue to exist; amodal completeness -- objects have 3D occupancy, even if only partial observations are available; spatiotemporal continuity -- the movement of each object is continuous over space and time. In this paper, we introduce 3D Dynamic Scene Representation (DSR), a 3D volumetric scene representation that simultaneously discovers, tracks, reconstructs objects, and predicts their dynamics while capturing all three properties. We further propose DSR-Net, which learns to aggregate visual observations over multiple interactions to gradually build and refine DSR. Our model achieves state-of-the-art performance in modeling 3D scene dynamics with DSR on both simulated and real data. Combined with model predictive control, DSR-Net enables accurate planning in downstream robotic manipulation tasks such as planar pushing. Video is available at https://youtu.be/GQjYG3nQJ80.","url_abs":"https://arxiv.org/abs/2011.01968v2","url_pdf":"https://arxiv.org/pdf/2011.01968v2.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-3d-dynamic-scene-representations-for","repo_url":"https://github.com/columbia-robovision/dsr","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-3d-dynamic-scene-representations-for","repo_url":"https://github.com/columbia-ai-robotics/dsr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"model-predictive-control","task_name":"Model Predictive Control"},{"task_slug":"robot-manipulation","task_name":"Robot Manipulation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2011.01968","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}