{"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/exploring-correlations-in-multiple-facial","title":"Exploring Correlations in Multiple Facial Attributes through Graph Attention Network","arxiv_id":"1810.09162","date":"2018-10-22","proceeding":null,"authors":["Yan Zhang","Li Sun"],"abstract":"Estimating multiple attributes from a single facial image gives comprehensive\ndescriptions on the high level semantics of the face. It is naturally regarded\nas a multi-task supervised learning problem with a single deep CNN, in which\nlower layers are shared, and higher ones are task-dependent with the\nmulti-branch structure. Within the traditional deep multi-task learning (DMTL)\nframework, this paper intends to fully exploit the correlations among different\nattributes by constructing a graph. The node in graph represents the feature\nvector from a particular branch for a given attribute, and the edge can be\ndefined by either the prior knowledge or the similarity between two nodes in\nthe embedding with a fully data-driven manner. We analyze that the attention\nmechanism actually takes effect in the latter case, and utilize the Graph\nAttention Layer (GAL) for exploring on the most relevant attribute feature and\nrefining the task-dependant feature by considering other attributes.\nExperiments show that by mining the correlations among attributes, our method\ncan improve the recognition accuracy on CelebA and LFWA dataset. And it also\nachieves competitive performance.","url_abs":"http://arxiv.org/abs/1810.09162v1","url_pdf":"http://arxiv.org/pdf/1810.09162v1.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":"exploring-correlations-in-multiple-facial","repo_url":"https://github.com/crazydemo/facial-attribute-classification-with-graph","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"graph-attention","task_name":"Graph Attention"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}