{"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/faceposenet-making-a-case-for-landmark-free","title":"FacePoseNet: Making a Case for Landmark-Free Face Alignment","arxiv_id":"1708.07517","date":"2017-08-24","proceeding":null,"authors":["Feng-Ju Chang","Anh Tuan Tran","Tal Hassner","Iacopo Masi","Ram Nevatia","Gerard Medioni"],"abstract":"We show how a simple convolutional neural network (CNN) can be trained to\naccurately and robustly regress 6 degrees of freedom (6DoF) 3D head pose,\ndirectly from image intensities. We further explain how this FacePoseNet (FPN)\ncan be used to align faces in 2D and 3D as an alternative to explicit facial\nlandmark detection for these tasks. We claim that in many cases the standard\nmeans of measuring landmark detector accuracy can be misleading when comparing\ndifferent face alignments. Instead, we compare our FPN with existing methods by\nevaluating how they affect face recognition accuracy on the IJB-A and IJB-B\nbenchmarks: using the same recognition pipeline, but varying the face alignment\nmethod. Our results show that (a) better landmark detection accuracy measured\non the 300W benchmark does not necessarily imply better face recognition\naccuracy. (b) Our FPN provides superior 2D and 3D face alignment on both\nbenchmarks. Finally, (c), FPN aligns faces at a small fraction of the\ncomputational cost of comparably accurate landmark detectors. For many\npurposes, FPN is thus a far faster and far more accurate face alignment method\nthan using facial landmark detectors.","url_abs":"http://arxiv.org/abs/1708.07517v2","url_pdf":"http://arxiv.org/pdf/1708.07517v2.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":"faceposenet-making-a-case-for-landmark-free","repo_url":"https://github.com/fengju514/Face-Pose-Net","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"faceposenet-making-a-case-for-landmark-free","repo_url":"https://github.com/fengju514/Expression-Net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"faceposenet-making-a-case-for-landmark-free","repo_url":"https://github.com/iacopomasi/face_specific_augm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"faceposenet-making-a-case-for-landmark-free","repo_url":"https://github.com/nova26/facePoseEstimation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"faceposenet-making-a-case-for-landmark-free","repo_url":"https://github.com/Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"3d-face-alignment","task_name":"3D Face Alignment"},{"task_slug":"face-alignment","task_name":"Face Alignment"},{"task_slug":"face-identification","task_name":"Face Identification"},{"task_slug":"face-recognition","task_name":"Face Recognition"},{"task_slug":"face-verification","task_name":"Face Verification"},{"task_slug":"facial-landmark-detection","task_name":"Facial Landmark Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-identification-on-ijb-a","task":"Face Identification","dataset":"IJB-A","model":"FPN","rank_in_archive_order":3,"of":3,"metrics":{"Accuracy":"91.4%"},"uses_additional_data":false},{"leaderboard":"/sota/face-identification-on-ijb-b","task":"Face Identification","dataset":"IJB-B","model":"FPN","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"91.1%"},"uses_additional_data":false},{"leaderboard":"/sota/face-verification-on-ijb-a","task":"Face Verification","dataset":"IJB-A","model":"FPN","rank_in_archive_order":11,"of":17,"metrics":{"TAR @ FAR=0.01":"90.1%"},"uses_additional_data":false},{"leaderboard":"/sota/face-verification-on-ijb-b","task":"Face Verification","dataset":"IJB-B","model":"FPN","rank_in_archive_order":5,"of":12,"metrics":{"TAR @ FAR=0.01":"96.5%"},"uses_additional_data":false},{"leaderboard":"/sota/facial-landmark-detection-on-300w","task":"Facial Landmark Detection","dataset":"300W","model":"FPN","rank_in_archive_order":15,"of":15,"metrics":{"Mean Error Rate":"0.1043"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1708.07517","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}