{"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/arbitrary-style-transfer-with-deep-feature","title":"Arbitrary Style Transfer with Deep Feature Reshuffle","arxiv_id":"1805.04103","date":"2018-05-10","proceeding":"CVPR 2018 6","authors":["Shuyang Gu","Congliang Chen","Jing Liao","Lu Yuan"],"abstract":"This paper introduces a novel method by reshuffling deep features (i.e.,\npermuting the spacial locations of a feature map) of the style image for\narbitrary style transfer. We theoretically prove that our new style loss based\non reshuffle connects both global and local style losses respectively used by\nmost parametric and non-parametric neural style transfer methods. This simple\nidea can effectively address the challenging issues in existing style transfer\nmethods. On one hand, it can avoid distortions in local style patterns, and\nallow semantic-level transfer, compared with neural parametric methods. On the\nother hand, it can preserve globally similar appearance to the style image, and\navoid wash-out artifacts, compared with neural non-parametric methods. Based on\nthe proposed loss, we also present a progressive feature-domain optimization\napproach. The experiments show that our method is widely applicable to various\nstyles, and produces better quality than existing methods.","url_abs":"http://arxiv.org/abs/1805.04103v4","url_pdf":"http://arxiv.org/pdf/1805.04103v4.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":"arbitrary-style-transfer-with-deep-feature","repo_url":"https://github.com/msracver/Style-Feature-Reshuffle","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"style-transfer","task_name":"Style Transfer"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.04103","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}