[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83198-en":3,"doc-seo-83198-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},83198,1374391974468,"Eden","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","ColorFM An Optimization to Learning Framework for Color Transfer via Flow Matching","Color transfer aligns a source image’s color distribution to a reference while preserving structure and semantics, yet prior work often causes inaccurate global mapping, semantic misalignment, and visual artifacts. ColorFM introduces an optimization-to-learning framework that reformulates color transfer as transporting pixel distributions along velocity fields via Flow Matching. ColorFM-O performs instance-specific hierarchical color coupling with semantic priors and produces pseudo-supervised pairs from integrated flow trajectories. ColorFM-L trains a feed-forward model to predict flow parameters using implicit state modeling, yielding high-quality, semantically consistent results faster than optimization.","arXiv :2607 .07 1 19v 1 [ cs .CV] 8 Jul 2026  \nColorFM: An Optimization-to-Learning Framework for Color Transfer via Flow Matching  \nYuhang He 1 , Kai Zhang 1 B, Xiaoming Li 1 , Du Chen2 , and Jian Yang 1  \n1 School of Intelligence Science and Technology, Nanjing University, China  \n2 VIVO BlueImage Lab, China  \n[https://github.com/cszn/ColorFM](https://github.com/cszn/ColorFM)  \nAbstract. Color transfer aims to align the color distribution of a source image with that of a reference image while preserving structural and semantic consistency. However, existing methods often suffer from inaccurate global mapping, semantic misalignment, and visual artifacts. To address these issues, we propose ColorFM, an optimization-to-learning framework. ColorFM connects online optimization to offline inference by reformulating color transfer as the transport of pixel distributions along velocity fields via Flow Matching. Specifically, we introduce ColorFMO, an instance-specific optimization scheme that fits the velocity field through hierarchical color coupling guided by semantic priors. By numerically integrating the induced flow trajectories, ColorFM-O produces precise and semantically consistent color transfer results, while generating high-quality paired data as pseudo-supervision. Building upon this, we design ColorFM-L, an efficient feed-forward model trained on the generated pairs. Through implicit state modeling, ColorFM-L extracts deep semantic features to predict flow parameters for bidirectional linearized transport, ensuring accurate color transfer. Extensive experiments demonstrate that ColorFM-L outperforms state-of-the-art methods in visual quality, structural fidelity, and semantic consistency, successfully combining the accuracy of optimization with the speed of feedforward inference.  \nKeywords: Color Transfer · Flow Matching · Optimization-to-Learning  \n1 Introduction  \nColor retouching plays a crucial role in visual communication and digital photography, as it shapes the perceived style and atmosphere of an image. However, the strong interdependence among color channels makes precise adjustment difficult for non-experts. While traditional approaches like image filters and Look-Up Tables (LUTs) offer simplified solutions, they typically apply rigid transformations that disregard the intrinsic color distribution of the source image, often lead  \ning to suboptimal results. To overcome these limitations, many automated color transfer techniques have been proposed. Given a reference style image, these B Corresponding author ([kaizhang@nju.edu.cn](kaizhang@nju.edu.cn))  \n2 Y. He et al.  \nStyle  \nContent Results of ColorFM-L on different styles  \nFig. 1: Color Transfer Results of ColorFM-L. ColorFM-L enables precise color transfer across diverse styles while preserving structural fidelity and semantic consistency.  \nmethods aim to transfer its color style to the content image while maintaining photorealism.  \nBroadly, existing methods can be categorized into two paradigms: online optimization and offline inference. Optimization-based approaches [4, 17 , 25 , 28–30] are often instance-specific and computationally intensive. Moreover, their reliance on hand-crafted or imperfect optimization objectives frequently results in inaccurate color transfer. Conversely, offline techniques are predominantly learning-based [3, 5 , 10–12, 14 , 16 , 19 , 20 , 40 , 41 , 44] . Feature-transformation-based methods [5, 19 , 40 , 44] match deep statistics between images but often introduce visual artifacts or severe color banding. Mapping-based approaches [10, 14 , 20](e.g ., LUT generation [22,45]) learn direct color transformations, yet their performance is limited by data scarcity and synthetic dataset biases, leading to poor generalization in complex real-world scenes. Furthermore, a recent flow-based method [16] focuses on learning global distribution alignment, often overlooking image content, which degrades visual quality.  \nBeyond these spec","cbCaip1JSO8f0mUC","https://ap.wps.com/l/cbCaip1JSO8f0mUC","pdf",21182144,1,17,"English","en",105,"# Introduction\n# Related Work\n## Color Transfer","[{\"question\":\"What problem does ColorFM address in existing color transfer methods?\",\"answer\":\"Existing methods often produce inaccurate global color mapping, semantic misalignment, and visual artifacts. ColorFM targets these issues by reformulating the task using Flow Matching and semantic-guided optimization.\"},{\"question\":\"How does ColorFM model color transfer in its optimization-to-learning framework?\",\"answer\":\"ColorFM treats color transfer as transporting pixel distributions in color space along velocity fields, implemented via Flow Matching. It connects online optimization to offline inference through this shared formulation.\"},{\"question\":\"What are the roles of ColorFM-O and ColorFM-L?\",\"answer\":\"ColorFM-O is an instance-specific optimization scheme that fits velocity fields using hierarchical color coupling guided by semantic priors and generates pseudo-supervised paired data. ColorFM-L is a feed-forward model trained on those pairs to predict flow parameters efficiently for accurate color transfer.\"}]",1784185904,43,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"colorfm-an-optimization-to-learning-framework-for-color-transfer-via-flow-matching","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/colorfm-an-optimization-to-learning-framework-for-color-transfer-via-flow-matching/83198/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does ColorFM address in existing color transfer methods?","Question",{"text":75,"@type":76},"Existing methods often produce inaccurate global color mapping, semantic misalignment, and visual artifacts. ColorFM targets these issues by reformulating the task using Flow Matching and semantic-guided optimization.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does ColorFM model color transfer in its optimization-to-learning framework?",{"text":80,"@type":76},"ColorFM treats color transfer as transporting pixel distributions in color space along velocity fields, implemented via Flow Matching. It connects online optimization to offline inference through this shared formulation.",{"name":82,"@type":73,"acceptedAnswer":83},"What are the roles of ColorFM-O and ColorFM-L?",{"text":84,"@type":76},"ColorFM-O is an instance-specific optimization scheme that fits velocity fields using hierarchical color coupling guided by semantic priors and generates pseudo-supervised paired data. ColorFM-L is a feed-forward model trained on those pairs to predict flow parameters efficiently for accurate color transfer.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]