{"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/fiducial-focus-augmentation-for-facial","title":"Fiducial Focus Augmentation for Facial Landmark Detection","arxiv_id":"2402.15044","date":"2024-02-23","proceeding":null,"authors":["Purbayan Kar","Vishal Chudasama","Naoyuki Onoe","Pankaj Wasnik","Vineeth Balasubramanian"],"abstract":"Deep learning methods have led to significant improvements in the performance on the facial landmark detection (FLD) task. However, detecting landmarks in challenging settings, such as head pose changes, exaggerated expressions, or uneven illumination, continue to remain a challenge due to high variability and insufficient samples. This inadequacy can be attributed to the model's inability to effectively acquire appropriate facial structure information from the input images. To address this, we propose a novel image augmentation technique specifically designed for the FLD task to enhance the model's understanding of facial structures. To effectively utilize the newly proposed augmentation technique, we employ a Siamese architecture-based training mechanism with a Deep Canonical Correlation Analysis (DCCA)-based loss to achieve collective learning of high-level feature representations from two different views of the input images. Furthermore, we employ a Transformer + CNN-based network with a custom hourglass module as the robust backbone for the Siamese framework. Extensive experiments show that our approach outperforms multiple state-of-the-art approaches across various benchmark datasets.","url_abs":"https://arxiv.org/abs/2402.15044v1","url_pdf":"https://arxiv.org/pdf/2402.15044v1.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":[],"tasks":[{"task_slug":"face-alignment","task_name":"Face Alignment"},{"task_slug":"facial-landmark-detection","task_name":"Facial Landmark Detection"},{"task_slug":"image-augmentation","task_name":"Image Augmentation"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"hourglass-module","method_name":"Hourglass Module"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-alignment-on-300w","task":"Face Alignment","dataset":"300W","model":"FiFA","rank_in_archive_order":4,"of":48,"metrics":{"NME_inter-ocular (%, Challenge)":"4.47","NME_inter-ocular (%, Common)":"2.51","NME_inter-ocular (%, Full)":"2.89"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-aflw-19","task":"Face Alignment","dataset":"AFLW-19","model":"FiFA","rank_in_archive_order":1,"of":23,"metrics":{"AUC_box@0.07 (%, Full)":"81.8","NME_box (%, Full)":"1.31","NME_diag (%, Frontal)":"0.80","NME_diag (%, Full)":"0.92"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-cofw","task":"Face Alignment","dataset":"COFW","model":"FiFA","rank_in_archive_order":1,"of":28,"metrics":{"NME (inter-ocular)":"2.96"},"uses_additional_data":false},{"leaderboard":"/sota/facial-landmark-detection-on-300w","task":"Facial Landmark Detection","dataset":"300W","model":"FiFA","rank_in_archive_order":2,"of":15,"metrics":{"NME":"2.89"},"uses_additional_data":false},{"leaderboard":"/sota/facial-landmark-detection-on-aflw-front","task":"Facial Landmark Detection","dataset":"AFLW-Front","model":"FiFA","rank_in_archive_order":1,"of":3,"metrics":{"Mean NME":"0.80","Mean NME ":"0.80","NME":"0.80"},"uses_additional_data":false},{"leaderboard":"/sota/facial-landmark-detection-on-aflw-full","task":"Facial Landmark Detection","dataset":"AFLW-Full","model":"FiFA","rank_in_archive_order":1,"of":5,"metrics":{"Mean NME":"0.92","Mean NME ":"0.92","NME":"0.92"},"uses_additional_data":false},{"leaderboard":"/sota/facial-landmark-detection-on-cofw","task":"Facial Landmark Detection","dataset":"COFW","model":"FiFA","rank_in_archive_order":2,"of":2,"metrics":{"NME":"2.96","NME (inter-ocular)":"2.96"},"uses_additional_data":false},{"leaderboard":"/sota/facial-landmark-detection-on-wflw-1","task":"Facial Landmark Detection","dataset":"WFLW","model":"FiFA","rank_in_archive_order":3,"of":3,"metrics":{"AUC@10 (inter-ocular)":"61.78","FR@10 (inter-ocular)":"1.60","NME (inter-ocular)":"3.89"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}