{"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/modular-generative-adversarial-networks","title":"Modular Generative Adversarial Networks","arxiv_id":"1804.03343","date":"2018-04-10","proceeding":"ECCV 2018 9","authors":["Bo Zhao","Bo Chang","Zequn Jie","Leonid Sigal"],"abstract":"Existing methods for multi-domain image-to-image translation (or generation)\nattempt to directly map an input image (or a random vector) to an image in one\nof the output domains. However, most existing methods have limited scalability\nand robustness, since they require building independent models for each pair of\ndomains in question. This leads to two significant shortcomings: (1) the need\nto train exponential number of pairwise models, and (2) the inability to\nleverage data from other domains when training a particular pairwise mapping.\nInspired by recent work on module networks, this paper proposes ModularGAN for\nmulti-domain image generation and image-to-image translation. ModularGAN\nconsists of several reusable and composable modules that carry on different\nfunctions (e.g., encoding, decoding, transformations). These modules can be\ntrained simultaneously, leveraging data from all domains, and then combined to\nconstruct specific GAN networks at test time, according to the specific image\ntranslation task. This leads to ModularGAN's superior flexibility of generating\n(or translating to) an image in any desired domain. Experimental results\ndemonstrate that our model not only presents compelling perceptual results but\nalso outperforms state-of-the-art methods on multi-domain facial attribute\ntransfer.","url_abs":"http://arxiv.org/abs/1804.03343v1","url_pdf":"http://arxiv.org/pdf/1804.03343v1.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":"modular-generative-adversarial-networks","repo_url":"https://github.com/Tian-Jiang/ModularGAN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"modular-generative-adversarial-networks","repo_url":"https://github.com/lucasbotang/modulargan","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.03343","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}