{"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/when-computer-vision-gazes-at-cognition","title":"When Computer Vision Gazes at Cognition","arxiv_id":"1412.2672","date":"2014-12-08","proceeding":null,"authors":["Tao Gao","Daniel Harari","Joshua Tenenbaum","Shimon Ullman"],"abstract":"Joint attention is a core, early-developing form of social interaction. It is\nbased on our ability to discriminate the third party objects that other people\nare looking at. While it has been shown that people can accurately determine\nwhether another person is looking directly at them versus away, little is known\nabout human ability to discriminate a third person gaze directed towards\nobjects that are further away, especially in unconstraint cases where the\nlooker can move her head and eyes freely. In this paper we address this\nquestion by jointly exploring human psychophysics and a cognitively motivated\ncomputer vision model, which can detect the 3D direction of gaze from 2D face\nimages. The synthesis of behavioral study and computer vision yields several\ninteresting discoveries. (1) Human accuracy of discriminating targets\n8{\\deg}-10{\\deg} of visual angle apart is around 40% in a free looking gaze\ntask; (2) The ability to interpret gaze of different lookers vary dramatically;\n(3) This variance can be captured by the computational model; (4) Human\noutperforms the current model significantly. These results collectively show\nthat the acuity of human joint attention is indeed highly impressive, given the\ncomputational challenge of the natural looking task. Moreover, the gap between\nhuman and model performance, as well as the variability of gaze interpretation\nacross different lookers, require further understanding of the underlying\nmechanisms utilized by humans for this challenging task.","url_abs":"http://arxiv.org/abs/1412.2672v1","url_pdf":"http://arxiv.org/pdf/1412.2672v1.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":"when-computer-vision-gazes-at-cognition","repo_url":"https://github.com/horanyinora/gazeworkshop.github.io","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"task-2","task_name":"Task 2"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}