{"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/separating-style-and-content-for-generalized","title":"Separating Style and Content for Generalized Style Transfer","arxiv_id":"1711.06454","date":"2017-11-17","proceeding":"CVPR 2018 6","authors":["Yexun Zhang","Ya zhang","Wenbin Cai","Jie Chang"],"abstract":"Neural style transfer has drawn broad attention in recent years. However,\nmost existing methods aim to explicitly model the transformation between\ndifferent styles, and the learned model is thus not generalizable to new\nstyles. We here attempt to separate the representations for styles and\ncontents, and propose a generalized style transfer network consisting of style\nencoder, content encoder, mixer and decoder. The style encoder and content\nencoder are used to extract the style and content factors from the style\nreference images and content reference images, respectively. The mixer employs\na bilinear model to integrate the above two factors and finally feeds it into a\ndecoder to generate images with target style and content. To separate the style\nfeatures and content features, we leverage the conditional dependence of styles\nand contents given an image. During training, the encoder network learns to\nextract styles and contents from two sets of reference images in limited size,\none with shared style and the other with shared content. This learning\nframework allows simultaneous style transfer among multiple styles and can be\ndeemed as a special `multi-task' learning scenario. The encoders are expected\nto capture the underlying features for different styles and contents which is\ngeneralizable to new styles and contents. For validation, we applied the\nproposed algorithm to the Chinese Typeface transfer problem. Extensive\nexperiment results on character generation have demonstrated the effectiveness\nand robustness of our method.","url_abs":"http://arxiv.org/abs/1711.06454v6","url_pdf":"http://arxiv.org/pdf/1711.06454v6.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":"separating-style-and-content-for-generalized","repo_url":"https://github.com/ycjing/Character-Stylization","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"style-transfer","task_name":"Style Transfer"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.06454","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}