[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-272337-105":59,"doc-detail-272337-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","ssat-a-symmetric-semantic-aware-transformer-network-for-makeup-transfer-and-removal-paper-abstract","SSAT - A Symmetric Semantic-Aware Transformer Network for Makeup Transfer and Removal - Paper Abstract","","Makeup transfer should not only extract makeup style from a reference image, but also apply it to the semantic matching regions of a target face. Existing approaches emphasize style extraction while overlooking semantic correspondence, which causes misaligned makeup, color averaging, and degraded robustness to expression, pose, and occlusion. SSAT proposes a unified Symmetric Semantic-Aware Transformer with a symmetric SSCFT feature transfer module and a weakly supervised semantic loss. Experiments demonstrate visually superior transfer results, stronger user-study preference, and improved robustness, including extension to video makeup transfer.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/ssat-a-symmetric-semantic-aware-transformer-network-for-makeup-transfer-and-removal-paper-abstract/272337/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/ssat-a-symmetric-semantic-aware-transformer-network-for-makeup-transfer-and-removal-paper-abstract/272337.png","ImageObject",300,407,{"name":92,"@type":93},"Olivia Brown","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-17","2026-09-15",true,{"@type":102,"interactionType":103,"userInteractionCount":29},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does SSAT address in makeup transfer?","Question",{"text":112,"@type":113},"SSAT targets the lack of semantic correspondence in existing methods, which leads to makeup being applied to incorrect facial regions and to visually inaccurate color and robustness outcomes.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does SSAT enforce semantic alignment between reference and target images?",{"text":117,"@type":113},"SSAT introduces a Symmetric Semantic Corresponding Feature Transfer (SSCFT) module to spatially distort makeup features for semantic alignment, supported by a weakly supervised semantic loss.",{"name":119,"@type":110,"acceptedAnswer":120},"What evidence supports the effectiveness of the proposed method?",{"text":121,"@type":113},"Experiments show more visually accurate makeup transfer results, superiority in user studies compared with state-of-the-art methods, and robustness across expression/pose, object occlusion scenes, and video makeup transfer.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},272337,1789464453,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":29,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":52,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":129,"read_time":143,"preview_page_count":4},16904993612988,"https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd","SSAT: A Symmetric Semantic-Aware Transformer Network for Makeup Transfer  \nand Removal  \nZhaoyang Sun 1 , Yaxiong Chen 1, 2 , Shengwu Xiong 1, 3, 􀀃  \n1 School of Computer Science and Arti􀀂cial Intelligence, Wuhan University of Technology, Wuhan, 430070  \n2 Wuhan University of Technology Chongqing Research Institute, Chongqing, 401122  \n3 Sanya Science and Education Innovation Park of Wuhan University of Technology, Sanya, 572000  \n[zhaoyangsun0304@outlook.com](zhaoyangsun0304@outlook.com), [chenyaxiong@whut.edu.cn](chenyaxiong@whut.edu.cn), [xiongsw@whut.edu.cn](xiongsw@whut.edu.cn)  \nAbstract  \nMakeup transfer is not only to extract the makeup style of the reference image, but also to render the makeup style to the semantic corresponding position of the target image. However, most existing methods focus on the former and ignore the latter, resulting in a failure to achieve desired results. To solve the above problems, we propose a uni􀀂ed Symmetric Semantic-Aware Transformer (SSAT) network, which incorporates semantic correspondence learning to realize makeup transfer and removal simultaneously. In SSAT, a novel Symmetric Semantic Corresponding Feature Transfer (SSCFT) module and a weakly supervised semantic loss are proposed to model and facilitate the establishment of accurate semantic correspondence. In the generation process, the extracted makeup features are spatially distorted by SSCFT to achieve semantic alignment with the target image, then the distorted makeup features are combined with unmodi􀀂ed makeup irrelevant features to produce the 􀀂nal result. Experiments show that our method obtains more visually accurate makeup transfer results, and user study in comparison with other state-ofthe-art makeup transfer methods re􀀃ects the superiority of our method. Besides, we verify the robustness of the proposed method in the difference of expression and pose, object occlusion scenes, and extend it to video makeup transfer.  \nIntroduction  \nMakeup transfer aims to transfer the makeup style of any reference image to the target image while preserving the identity of the target person. With the booming development of the cosmetics market, makeup transfer is widely demanded in many popular beautifying applications and has been received extensive attention in the computer vision and graphics. In the early years, traditional approaches (Guo and Sim 2009; Tong et al. 2007; Li, Zhou, and Lin 2015) mostly used image gradient editing or physical-based modi􀀂cation to realize makeup transfer. Recently, combining generative adversarial network (Goodfellow et al. 2014) and disentangled representation (Lee et al. 2018; Huang et al. 2018), many makeup transfer approaches (Li et al. 2018; Chang et al. 2018; Chen et al. 2019; Gu et al. 2019; Huang et al. 2020; Deng et al. 202 1; Nguyen, Tran, and Hoai 2021) have made  \n􀀃 Corresponding Author  \nCopyright 􀀍c 2022, Association for the Advancement of Arti􀀂cial Intelligence ([www.aaai.org](www.aaai.org)). All rights reserved.  \nFigure 1: The diagram of our motivation and PSGAN comparison. The spatial mapping of the same color points between the target image and the reference image is the semantic correspondence expected to be established in this paper.  \nsigni􀀂cant progress in extracting makeup style and generating realistic makeup results.  \nHowever, little attention has been paid to semantic correspondence, which plays an important role in makeup transfer. Consider the actual makeup process following the tutorial: 1) choose cosmetics of the same color (extract makeup style); 2) apply this cosmetics to the semantic corresponding position of the face (semantic correspondence), see Figure 1 . Ignoring the semantic correspondence would result in the makeup style being transferred to the wrong position, such as lipstick leaking lips. At the same time, inaccurate semantic correspondence will lead to the averaging of makeup colors (Jiang et al. 2020), so that the resulting makeup style is visually differ","cbCaifnEtNVAX5Ww","https://ap.wps.com/l/cbCaifnEtNVAX5Ww","pdf",13918490,"English","# Abstract\n# Introduction\n## Motivation and Problem Setting\n## Proposed Solution: SSAT and SSCFT\n## Contributions and Differences","[{\"question\":\"What problem does SSAT address in makeup transfer?\",\"answer\":\"SSAT targets the lack of semantic correspondence in existing methods, which leads to makeup being applied to incorrect facial regions and to visually inaccurate color and robustness outcomes.\"},{\"question\":\"How does SSAT enforce semantic alignment between reference and target images?\",\"answer\":\"SSAT introduces a Symmetric Semantic Corresponding Feature Transfer (SSCFT) module to spatially distort makeup features for semantic alignment, supported by a weakly supervised semantic loss.\"},{\"question\":\"What evidence supports the effectiveness of the proposed method?\",\"answer\":\"Experiments show more visually accurate makeup transfer results, superiority in user studies compared with state-of-the-art methods, and robustness across expression/pose, object occlusion scenes, and video makeup transfer.\"}]","SSAT - A Symmetric Semantic-Aware Transformer Network for Makeup Transfer and Removal - Paper Abstract | PDF",25]