{"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/multimodal-transfer-a-hierarchical-deep","title":"Multimodal Transfer: A Hierarchical Deep Convolutional Neural Network for Fast Artistic Style Transfer","arxiv_id":"1612.01895","date":"2016-11-17","proceeding":"CVPR 2017 7","authors":["Xin Wang","Geoffrey Oxholm","Da Zhang","Yuan-Fang Wang"],"abstract":"Transferring artistic styles onto everyday photographs has become an\nextremely popular task in both academia and industry. Recently, offline\ntraining has replaced on-line iterative optimization, enabling nearly real-time\nstylization. When those stylization networks are applied directly to\nhigh-resolution images, however, the style of localized regions often appears\nless similar to the desired artistic style. This is because the transfer\nprocess fails to capture small, intricate textures and maintain correct texture\nscales of the artworks. Here we propose a multimodal convolutional neural\nnetwork that takes into consideration faithful representations of both color\nand luminance channels, and performs stylization hierarchically with multiple\nlosses of increasing scales. Compared to state-of-the-art networks, our network\ncan also perform style transfer in nearly real-time by conducting much more\nsophisticated training offline. By properly handling style and texture cues at\nmultiple scales using several modalities, we can transfer not just large-scale,\nobvious style cues but also subtle, exquisite ones. That is, our scheme can\ngenerate results that are visually pleasing and more similar to multiple\ndesired artistic styles with color and texture cues at multiple scales.","url_abs":"http://arxiv.org/abs/1612.01895v2","url_pdf":"http://arxiv.org/pdf/1612.01895v2.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":"multimodal-transfer-a-hierarchical-deep","repo_url":"https://github.com/FeliMe/multimodal_style_transfer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"multimodal-transfer-a-hierarchical-deep","repo_url":"https://github.com/fullfanta/multimodal_transfer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","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=1612.01895","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}