{"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/multi-task-convolutional-neural-network-for","title":"Multi-Task Convolutional Neural Network for Pose-Invariant Face Recognition","arxiv_id":"1702.04710","date":"2017-02-15","proceeding":null,"authors":["Xi Yin","Xiaoming Liu"],"abstract":"This paper explores multi-task learning (MTL) for face recognition. We answer\nthe questions of how and why MTL can improve the face recognition performance.\nFirst, we propose a multi-task Convolutional Neural Network (CNN) for face\nrecognition where identity classification is the main task and pose,\nillumination, and expression estimations are the side tasks. Second, we develop\na dynamic-weighting scheme to automatically assign the loss weight to each side\ntask, which is a crucial problem in MTL. Third, we propose a pose-directed\nmulti-task CNN by grouping different poses to learn pose-specific identity\nfeatures, simultaneously across all poses. Last but not least, we propose an\nenergy-based weight analysis method to explore how CNN-based MTL works. We\nobserve that the side tasks serve as regularizations to disentangle the\nvariations from the learnt identity features. Extensive experiments on the\nentire Multi-PIE dataset demonstrate the effectiveness of the proposed\napproach. To the best of our knowledge, this is the first work using all data\nin Multi-PIE for face recognition. Our approach is also applicable to\nin-the-wild datasets for pose-invariant face recognition and achieves\ncomparable or better performance than state of the art on LFW, CFP, and IJB-A\ndatasets.","url_abs":"http://arxiv.org/abs/1702.04710v2","url_pdf":"http://arxiv.org/pdf/1702.04710v2.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":"multi-task-convolutional-neural-network-for","repo_url":"https://github.com/xiyinmsu/MultiTask-CNN","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"face-recognition","task_name":"Face Recognition"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"robust-face-recognition","task_name":"Robust Face Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1702.04710","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}