{"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/sliced-wasserstein-generative-models","title":"Sliced Wasserstein Generative Models","arxiv_id":"1706.02631","date":"2017-06-08","proceeding":null,"authors":["Jiqing Wu","Zhiwu Huang","Dinesh Acharya","Wen Li","Janine Thoma","Danda Pani Paudel","Luc van Gool"],"abstract":"In generative modeling, the Wasserstein distance (WD) has emerged as a useful\nmetric to measure the discrepancy between generated and real data\ndistributions. Unfortunately, it is challenging to approximate the WD of\nhigh-dimensional distributions. In contrast, the sliced Wasserstein distance\n(SWD) factorizes high-dimensional distributions into their multiple\none-dimensional marginal distributions and is thus easier to approximate. In\nthis paper, we introduce novel approximations of the primal and dual SWD.\nInstead of using a large number of random projections, as it is done by\nconventional SWD approximation methods, we propose to approximate SWDs with a\nsmall number of parameterized orthogonal projections in an end-to-end deep\nlearning fashion. As concrete applications of our SWD approximations, we design\ntwo types of differentiable SWD blocks to equip modern generative\nframeworks---Auto-Encoders (AE) and Generative Adversarial Networks (GAN). In\nthe experiments, we not only show the superiority of the proposed generative\nmodels on standard image synthesis benchmarks, but also demonstrate the\nstate-of-the-art performance on challenging high resolution image and video\ngeneration in an unsupervised manner.","url_abs":"http://arxiv.org/abs/1706.02631v4","url_pdf":"http://arxiv.org/pdf/1706.02631v4.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":"sliced-wasserstein-generative-models","repo_url":"https://github.com/musikisomorphie/swd","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"video-generation","task_name":"Video Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}