{"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-to-look-around-intelligently","title":"Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks","arxiv_id":"1709.00507","date":"2017-09-01","proceeding":"CVPR 2018 6","authors":["Dinesh Jayaraman","Kristen Grauman"],"abstract":"It is common to implicitly assume access to intelligently captured inputs\n(e.g., photos from a human photographer), yet autonomously capturing good\nobservations is itself a major challenge. We address the problem of learning to\nlook around: if a visual agent has the ability to voluntarily acquire new views\nto observe its environment, how can it learn efficient exploratory behaviors to\nacquire informative observations? We propose a reinforcement learning solution,\nwhere the agent is rewarded for actions that reduce its uncertainty about the\nunobserved portions of its environment. Based on this principle, we develop a\nrecurrent neural network-based approach to perform active completion of\npanoramic natural scenes and 3D object shapes. Crucially, the learned policies\nare not tied to any recognition task nor to the particular semantic content\nseen during training. As a result, 1) the learned \"look around\" behavior is\nrelevant even for new tasks in unseen environments, and 2) training data\nacquisition involves no manual labeling. Through tests in diverse settings, we\ndemonstrate that our approach learns useful generic policies that transfer to\nnew unseen tasks and environments. Completion episodes are shown at\nhttps://goo.gl/BgWX3W.","url_abs":"http://arxiv.org/abs/1709.00507v2","url_pdf":"http://arxiv.org/pdf/1709.00507v2.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-to-look-around-intelligently","repo_url":"https://github.com/dineshj1/lookaround","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"unanswered"}},{"paper_slug":"learning-to-look-around-intelligently","repo_url":"https://github.com/srama2512/sidekicks","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1709.00507","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}