{"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/informative-object-annotations-tell-me","title":"Informative Object Annotations: Tell Me Something I Don't Know","arxiv_id":"1812.10358","date":"2018-12-26","proceeding":"CVPR 2019 6","authors":["Lior Bracha","Gal Chechik"],"abstract":"Capturing the interesting components of an image is a key aspect of image\nunderstanding. When a speaker annotates an image, selecting labels that are\ninformative greatly depends on the prior knowledge of a prospective listener.\nMotivated by cognitive theories of categorization and communication, we present\na new unsupervised approach to model this prior knowledge and quantify the\ninformativeness of a description. Specifically, we compute how knowledge of a\nlabel reduces uncertainty over the space of labels and utilize this to rank\ncandidate labels for describing an image. While the full estimation problem is\nintractable, we describe an efficient algorithm to approximate entropy\nreduction using a tree-structured graphical model. We evaluate our approach on\nthe open-images dataset using a new evaluation set of 10K ground-truth ratings\nand find that it achieves ~65% agreement with human raters, largely\noutperforming other unsupervised baseline approaches.","url_abs":"http://arxiv.org/abs/1812.10358v1","url_pdf":"http://arxiv.org/pdf/1812.10358v1.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":"informative-object-annotations-tell-me","repo_url":"https://github.com/liorbracha/iota","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"informativeness","task_name":"Informativeness"},{"task_slug":"object","task_name":"Object"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}