{"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/adviser-networks-learning-what-question-to","title":"Adviser Networks: Learning What Question to Ask for Human-In-The-Loop Viewpoint Estimation","arxiv_id":"1802.01666","date":"2018-02-05","proceeding":null,"authors":["Mohamed El Banani","Jason J. Corso"],"abstract":"Humans have an unparalleled visual intelligence and can overcome visual\nambiguities that machines currently cannot. Recent works have shown that\nincorporating guidance from humans during inference for monocular\nviewpoint-estimation can help overcome difficult cases in which the\ncomputer-alone would have otherwise failed. These hybrid intelligence\napproaches are hence gaining traction. However, deciding what question to ask\nthe human at inference time remains an unknown for these problems.\n  We address this question by formulating it as an Adviser Problem: can we\nlearn a mapping from the input to a specific question to ask the human to\nmaximize the expected positive impact to the overall task? We formulate a\nsolution to the adviser problem for viewpoint estimation using a deep network\nwhere the question asks for the location of a keypoint in the input image. We\nshow that by using the Adviser Network's recommendations, the model and the\nhuman outperforms the previous hybrid-intelligence state-of-the-art by 3.7%,\nand the computer-only state-of-the-art by 5.28% absolute.","url_abs":"http://arxiv.org/abs/1802.01666v3","url_pdf":"http://arxiv.org/pdf/1802.01666v3.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":"adviser-networks-learning-what-question-to","repo_url":"https://github.com/mbanani/adviser_networks","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"viewpoint-estimation","task_name":"Viewpoint Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1802.01666","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}