[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-272339-105":3,"doc-detail-272339-en":81,"detail-sidebar-cat-0-en-105":98},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","content-style-decoupling-for-unsupervised-makeup-transfer-without-generating-pseudo-ground-truth","Content-Style Decoupling for Unsupervised Makeup Transfer without Generating Pseudo Ground Truth","","Makeup transfer lacks real target images for supervision, making unsupervised training a central challenge. Existing approaches generate pseudo ground truths (PGTs), but inaccurate PGTs introduce supervision noise that degrades final transfer quality. The proposed Content-Style Decoupled Makeup Transfer (CSD-MT) avoids PGT generation entirely by decoupling face content and makeup style via frequency decomposition. It maximizes content and makeup consistency between transferred results and inputs using two new loss functions, validated through extensive quantitative and qualitative evaluation.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/content-style-decoupling-for-unsupervised-makeup-transfer-without-generating-pseudo-ground-truth/272339/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/content-style-decoupling-for-unsupervised-makeup-transfer-without-generating-pseudo-ground-truth/272339.png","ImageObject",300,407,{"name":42,"@type":43},"Xiajie","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-17","2026-09-15",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",13,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why is makeup transfer difficult in unsupervised settings?","Question",{"text":63,"@type":64},"Unsupervised makeup transfer has no real target images as supervision, so the model needs alternative guidance during training.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What is the main drawback of existing pseudo ground truth (PGT) based methods?",{"text":68,"@type":64},"The generated PGTs are often imprecise, and this imprecision acts as negative supervision, leading to performance degradation in the final transferred results.",{"name":70,"@type":61,"acceptedAnswer":71},"How does CSD-MT decouple content and makeup style information?",{"text":72,"@type":64},"CSD-MT uses frequency decomposition and assumes low-frequency components represent makeup style while high-frequency components preserve content details.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},272339,1789464461,{"code":4,"msg":82,"data":83},"success",{"doc_id":79,"user_id":84,"nickname":42,"user_avatar":85,"doc_module":4,"category_id":86,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":87,"file_id":88,"file_url":89,"file_type":90,"file_size":91,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":92,"language":93,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":94,"faqs":95,"seo_title":96,"seo_description":12,"update_tm":80,"read_time":97,"preview_page_count":4},8814010472675,"https://avatar.qwps.com/avatar/WGlhamll",8,"This CVPR paper is the Open Access version, provided by the Computer Vision Foundation.  \nExcept for this watermark, it is identical to the accepted version; the final published version of the proceedings is available on IEEE Xplore.  \nContent-Style Decoupling for Unsupervised Makeup Transfer without Generating Pseudo Ground Truth  \nZhaoyang Sun 1 Shengwu Xiong 1,2,3,4 Yaxiong Chen 1 Yi Rong 1 †  \n1Wuhan University of Technology 2Wuhan Huaxia Institute of Technology  \n3 Shanghai AI Laboratory 4 Qiongtai Normal University  \n[zhaoyangsun0304@outlook.com](zhaoyangsun0304@outlook.com), {xiongsw, chenyaxiong, [yrong](yrong}@whut.edu.cn)[}](yrong}@whut.edu.cn)[@whut.edu.cn](yrong}@whut.edu.cn)  \nAbstract  \nThe absence of real targets to guide the model training is one of the main problems with the makeup transfer task. Most existing methods tackle this problem by synthesizing pseudo ground truths (PGTs) . However, the generated PGTs are often sub-optimal and their imprecision will eventually lead to performance degradation. To alleviate this issue, in this paper, we propose a novel Content-Style Decoupled Makeup Transfer (CSD-MT) method, which works in a purely unsupervised manner and thus eliminates the negative effects of generating PGTs. Specifically, based on the frequency characteristics analysis, we assume that the low-frequency (LF) component of a face image is more associated with its makeup style information, while the high-frequency (HF) component is more related to its content details. This assumption allows CSDMT to decouple the content and makeup style information in each face image through the frequency decomposition. After that, CSD-MT realizes makeup transfer by maximizing the consistency of these two types of information between the transferred result and input images, respectively. Two newly designed loss functions are also introduced to further improve the transfer performance. Extensive quantitative and qualitative analyses show the effectiveness of our CSD-MT method. Our code is available at [https://github.com/Snowfallingplum/CSD-MT](https://github.com/Snowfallingplum/CSD-MT).  \n1. Introduction  \nGiven a pair of source and reference face images, the main goal of makeup transfer is to generate an image that simultaneously satisfies the following conditions: (1) Containing the makeup styles transferred from the reference image, such as lipstick, eye shadow and powder blush. (2) Preserving the content details of the source image, including iden-  \n†Corresponding author.  \ntity, facial structure and background. This technique has been widely studied and is attracting increasing attentions from the computer vision and artificial intelligence communities, due to its great economic potential in the fields of e-commerce, entertainment and beauty industries. However, considering the diversity and complexity of different makeup styles, makeup transfer remains a challenging task.  \nOne of the major problems with the makeup transfer task is its unsupervised nature, which means that there is no real transferred image that can be used as a target ground truth to guide the model training. To address this problem, most existing methods [3, 7, 9, 13, 17, 21, 23, 27, 30, 32] propose to synthesize a pseudo ground truth (PGT) image, from each input source-reference image pair, as an alternative supervision target. After that, the model parameters are optimized by minimizing the difference between each generated transferred output and its corresponding PGT (see Figure 1(a)) .  \nAs illustrated in Figure 2, according to the PGT generation strategy used, previous makeup transfer approaches can be roughly divided into two categories: (1) Histogrammatching-based methods [3, 9, 13, 15, 32] attempt to align the color distribution of each facial region (e.g., lip, eye and face areas) in the source image with that of the same region in the reference face. However, the PGT produced by this strategy discards all spatial information of makeup styl","cbCaidwyrbegccoR","https://ap.wps.com/l/cbCaidwyrbegccoR","pdf",5791822,10,"English","# Abstract\n# 1. Introduction\n## Unsupervised makeup transfer and pseudo ground truth issues\n## Categories of existing PGT generation strategies\n## Proposed CSD-MT approach","[{\"question\":\"Why is makeup transfer difficult in unsupervised settings?\",\"answer\":\"Unsupervised makeup transfer has no real target images as supervision, so the model needs alternative guidance during training.\"},{\"question\":\"What is the main drawback of existing pseudo ground truth (PGT) based methods?\",\"answer\":\"The generated PGTs are often imprecise, and this imprecision acts as negative supervision, leading to performance degradation in the final transferred results.\"},{\"question\":\"How does CSD-MT decouple content and makeup style information?\",\"answer\":\"CSD-MT uses frequency decomposition and assumes low-frequency components represent makeup style while high-frequency components preserve content details.\"}]","Content-Style Decoupling for Unsupervised Makeup Transfer without Generating Pseudo Ground Truth | PDF",25,{"code":4,"msg":82,"data":99},[100,104,108,112,117,122,127,130,135,138,141],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":101,"show_sort_weight":102,"slug":103},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":105,"show_sort_weight":106,"slug":107},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":109,"show_sort_weight":110,"slug":111},"Exam",70,"exam",{"id":113,"doc_module":4,"doc_module_name":25,"category_name":114,"show_sort_weight":115,"slug":116},5,"Comic",60,"comic",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":119,"show_sort_weight":120,"slug":121},6,"Technology",50,"technology",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":125,"slug":126},7,"Healthcare",40,"healthcare",{"id":86,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":128,"slug":129},30,"research-report",{"id":131,"doc_module":4,"doc_module_name":25,"category_name":132,"show_sort_weight":133,"slug":134},9,"Religion & Spirituality",20,"religion-spirituality",{"id":133,"doc_module":4,"doc_module_name":25,"category_name":136,"show_sort_weight":133,"slug":137},"World Cup","world-cup",{"id":92,"doc_module":4,"doc_module_name":25,"category_name":139,"show_sort_weight":92,"slug":140},"Lifestyle","lifestyle",{"id":142,"doc_module":4,"doc_module_name":25,"category_name":143,"show_sort_weight":113,"slug":144},19,"General","general"]