{"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/demystifying-neural-style-transfer","title":"Demystifying Neural Style Transfer","arxiv_id":"1701.01036","date":"2017-01-04","proceeding":null,"authors":["Yanghao Li","Naiyan Wang","Jiaying Liu","Xiaodi Hou"],"abstract":"Neural Style Transfer has recently demonstrated very exciting results which\ncatches eyes in both academia and industry. Despite the amazing results, the\nprinciple of neural style transfer, especially why the Gram matrices could\nrepresent style remains unclear. In this paper, we propose a novel\ninterpretation of neural style transfer by treating it as a domain adaptation\nproblem. Specifically, we theoretically show that matching the Gram matrices of\nfeature maps is equivalent to minimize the Maximum Mean Discrepancy (MMD) with\nthe second order polynomial kernel. Thus, we argue that the essence of neural\nstyle transfer is to match the feature distributions between the style images\nand the generated images. To further support our standpoint, we experiment with\nseveral other distribution alignment methods, and achieve appealing results. We\nbelieve this novel interpretation connects these two important research fields,\nand could enlighten future researches.","url_abs":"http://arxiv.org/abs/1701.01036v2","url_pdf":"http://arxiv.org/pdf/1701.01036v2.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":"demystifying-neural-style-transfer","repo_url":"https://github.com/aryan-mann/style-transfer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"demystifying-neural-style-transfer","repo_url":"https://github.com/sanjayjonckheere/iN_iS_Tee_One","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"demystifying-neural-style-transfer","repo_url":"https://github.com/sonnguyen129/deep-feature-rotation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"style-transfer","task_name":"Style Transfer"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1701.01036","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}