{"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/predicting-gaze-in-egocentric-video-by","title":"Predicting Gaze in Egocentric Video by Learning Task-dependent Attention Transition","arxiv_id":"1803.09125","date":"2018-03-24","proceeding":"ECCV 2018 9","authors":["Yifei Huang","Minjie Cai","Zhenqiang Li","Yoichi Sato"],"abstract":"We present a new computational model for gaze prediction in egocentric videos\nby exploring patterns in temporal shift of gaze fixations (attention\ntransition) that are dependent on egocentric manipulation tasks. Our assumption\nis that the high-level context of how a task is completed in a certain way has\na strong influence on attention transition and should be modeled for gaze\nprediction in natural dynamic scenes. Specifically, we propose a hybrid model\nbased on deep neural networks which integrates task-dependent attention\ntransition with bottom-up saliency prediction. In particular, the\ntask-dependent attention transition is learned with a recurrent neural network\nto exploit the temporal context of gaze fixations, e.g. looking at a cup after\nmoving gaze away from a grasped bottle. Experiments on public egocentric\nactivity datasets show that our model significantly outperforms\nstate-of-the-art gaze prediction methods and is able to learn meaningful\ntransition of human attention.","url_abs":"http://arxiv.org/abs/1803.09125v3","url_pdf":"http://arxiv.org/pdf/1803.09125v3.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":"predicting-gaze-in-egocentric-video-by","repo_url":"https://github.com/hyf015/egocentric-gaze-prediction","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"predicting-gaze-in-egocentric-video-by","repo_url":"https://github.com/hyf015/gaze_prediction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"eye-tracking","task_name":"Gaze Prediction"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"saliency-prediction","task_name":"Saliency Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.09125","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}