{"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/wasserstein-divergence-for-gans","title":"Wasserstein Divergence for GANs","arxiv_id":"1712.01026","date":"2017-12-04","proceeding":"ECCV 2018 9","authors":["Jiqing Wu","Zhiwu Huang","Janine Thoma","Dinesh Acharya","Luc van Gool"],"abstract":"In many domains of computer vision, generative adversarial networks (GANs)\nhave achieved great success, among which the family of Wasserstein GANs (WGANs)\nis considered to be state-of-the-art due to the theoretical contributions and\ncompetitive qualitative performance. However, it is very challenging to\napproximate the $k$-Lipschitz constraint required by the Wasserstein-1\nmetric~(W-met). In this paper, we propose a novel Wasserstein\ndivergence~(W-div), which is a relaxed version of W-met and does not require\nthe $k$-Lipschitz constraint. As a concrete application, we introduce a\nWasserstein divergence objective for GANs~(WGAN-div), which can faithfully\napproximate W-div through optimization. Under various settings, including\nprogressive growing training, we demonstrate the stability of the proposed\nWGAN-div owing to its theoretical and practical advantages over WGANs. Also, we\nstudy the quantitative and visual performance of WGAN-div on standard image\nsynthesis benchmarks of computer vision, showing the superior performance of\nWGAN-div compared to the state-of-the-art methods.","url_abs":"http://arxiv.org/abs/1712.01026v4","url_pdf":"http://arxiv.org/pdf/1712.01026v4.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":"wasserstein-divergence-for-gans","repo_url":"https://github.com/Lornatang/WassersteinGAN_DIV-PyTorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.01026","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}