{"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/reinforcement-learning-of-active-vision-for","title":"Reinforcement Learning of Active Vision for Manipulating Objects under Occlusions","arxiv_id":"1811.08067","date":"2018-11-20","proceeding":null,"authors":["Ricson Cheng","Arpit Agarwal","Katerina Fragkiadaki"],"abstract":"We consider artificial agents that learn to jointly control their gripperand\ncamera in order to reinforcement learn manipulation policies in the presenceof\nocclusions from distractor objects. Distractors often occlude the object of\nin-terest and cause it to disappear from the field of view. We propose hand/eye\ncon-trollers that learn to move the camera to keep the object within the field\nof viewand visible, in coordination to manipulating it to achieve the desired\ngoal, e.g.,pushing it to a target location. We incorporate structural biases of\nobject-centricattention within our actor-critic architectures, which our\nexperiments suggest tobe a key for good performance. Our results further\nhighlight the importance ofcurriculum with regards to environment difficulty.\nThe resulting active vision /manipulation policies outperform static camera\nsetups for a variety of clutteredenvironments.","url_abs":"http://arxiv.org/abs/1811.08067v2","url_pdf":"http://arxiv.org/pdf/1811.08067v2.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":"reinforcement-learning-of-active-vision-for","repo_url":"https://github.com/ricsonc/ActiveVisionManipulation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.08067","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}