{"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/beyond-textures-learning-from-multi-domain","title":"Learning from Multi-domain Artistic Images for Arbitrary Style Transfer","arxiv_id":"1805.09987","date":"2018-05-25","proceeding":null,"authors":["Zheng Xu","Michael Wilber","Chen Fang","Aaron Hertzmann","Hailin Jin"],"abstract":"We propose a fast feed-forward network for arbitrary style transfer, which\ncan generate stylized image for previously unseen content and style image\npairs. Besides the traditional content and style representation based on deep\nfeatures and statistics for textures, we use adversarial networks to regularize\nthe generation of stylized images. Our adversarial network learns the intrinsic\nproperty of image styles from large-scale multi-domain artistic images. The\nadversarial training is challenging because both the input and output of our\ngenerator are diverse multi-domain images. We use a conditional generator that\nstylized content by shifting the statistics of deep features, and a conditional\ndiscriminator based on the coarse category of styles. Moreover, we propose a\nmask module to spatially decide the stylization level and stabilize adversarial\ntraining by avoiding mode collapse. As a side effect, our trained discriminator\ncan be applied to rank and select representative stylized images. We\nqualitatively and quantitatively evaluate the proposed method, and compare with\nrecent style transfer methods. We release our code and model at\nhttps://github.com/nightldj/behance_release.","url_abs":"http://arxiv.org/abs/1805.09987v2","url_pdf":"http://arxiv.org/pdf/1805.09987v2.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":"beyond-textures-learning-from-multi-domain","repo_url":"https://github.com/nightldj/behance_release","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"style-transfer","task_name":"Style Transfer"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.09987","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}