{"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/island-loss-for-learning-discriminative","title":"Island Loss for Learning Discriminative Features in Facial Expression Recognition","arxiv_id":"1710.03144","date":"2017-10-09","proceeding":null,"authors":["Jie Cai","Zibo Meng","Ahmed Shehab Khan","Zhiyuan Li","James O'Reilly","Yan Tong"],"abstract":"Over the past few years, Convolutional Neural Networks (CNNs) have shown\npromise on facial expression recognition. However, the performance degrades\ndramatically under real-world settings due to variations introduced by subtle\nfacial appearance changes, head pose variations, illumination changes, and\nocclusions.\n  In this paper, a novel island loss is proposed to enhance the discriminative\npower of the deeply learned features. Specifically, the IL is designed to\nreduce the intra-class variations while enlarging the inter-class differences\nsimultaneously. Experimental results on four benchmark expression databases\nhave demonstrated that the CNN with the proposed island loss (IL-CNN)\noutperforms the baseline CNN models with either traditional softmax loss or the\ncenter loss and achieves comparable or better performance compared with the\nstate-of-the-art methods for facial expression recognition.","url_abs":"http://arxiv.org/abs/1710.03144v3","url_pdf":"http://arxiv.org/pdf/1710.03144v3.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":"island-loss-for-learning-discriminative","repo_url":"https://github.com/shanxuanchen/FacialExpressionRecognition","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"facial-expression-recognition-1","task_name":"Facial Expression Recognition"},{"task_slug":"facial-expression-recognition","task_name":"Facial Expression Recognition (FER)"}],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/facial-expression-recognition-on-sfew","task":"Facial Expression Recognition (FER)","dataset":"SFEW","model":"Island Loss","rank_in_archive_order":4,"of":4,"metrics":{"Accuracy":"52.52"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1710.03144","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}