{"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/quality-aware-network-for-set-to-set","title":"Quality Aware Network for Set to Set Recognition","arxiv_id":"1704.03373","date":"2017-04-11","proceeding":"CVPR 2017 7","authors":["Yu Liu","Junjie Yan","Wanli Ouyang"],"abstract":"This paper targets on the problem of set to set recognition, which learns the\nmetric between two image sets. Images in each set belong to the same identity.\nSince images in a set can be complementary, they hopefully lead to higher\naccuracy in practical applications. However, the quality of each sample cannot\nbe guaranteed, and samples with poor quality will hurt the metric. In this\npaper, the quality aware network (QAN) is proposed to confront this problem,\nwhere the quality of each sample can be automatically learned although such\ninformation is not explicitly provided in the training stage. The network has\ntwo branches, where the first branch extracts appearance feature embedding for\neach sample and the other branch predicts quality score for each sample.\nFeatures and quality scores of all samples in a set are then aggregated to\ngenerate the final feature embedding. We show that the two branches can be\ntrained in an end-to-end manner given only the set-level identity annotation.\nAnalysis on gradient spread of this mechanism indicates that the quality\nlearned by the network is beneficial to set-to-set recognition and simplifies\nthe distribution that the network needs to fit. Experiments on both face\nverification and person re-identification show advantages of the proposed QAN.\nThe source code and network structure can be downloaded at\nhttps://github.com/sciencefans/Quality-Aware-Network.","url_abs":"http://arxiv.org/abs/1704.03373v1","url_pdf":"http://arxiv.org/pdf/1704.03373v1.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":"quality-aware-network-for-set-to-set","repo_url":"https://github.com/sciencefans/Quality-Aware-Network","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"face-verification","task_name":"Face Verification"},{"task_slug":"person-re-identification","task_name":"Person Re-Identification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-verification-on-youtube-faces-db","task":"Face Verification","dataset":"YouTube Faces DB","model":"QAN","rank_in_archive_order":6,"of":12,"metrics":{"Accuracy":"96.17%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.03373","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}