{"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/towards-multi-pose-guided-virtual-try-on","title":"Towards Multi-pose Guided Virtual Try-on Network","arxiv_id":"1902.11026","date":"2019-02-28","proceeding":"ICCV 2019 10","authors":["Haoye Dong","Xiaodan Liang","Bochao Wang","Hanjiang Lai","Jia Zhu","Jian Yin"],"abstract":"Virtual try-on system under arbitrary human poses has huge application\npotential, yet raises quite a lot of challenges, e.g. self-occlusions, heavy\nmisalignment among diverse poses, and diverse clothes textures. Existing\nmethods aim at fitting new clothes into a person can only transfer clothes on\nthe fixed human pose, but still show unsatisfactory performances which often\nfail to preserve the identity, lose the texture details, and decrease the\ndiversity of poses. In this paper, we make the first attempt towards multi-pose\nguided virtual try-on system, which enables transfer clothes on a person image\nunder diverse poses. Given an input person image, a desired clothes image, and\na desired pose, the proposed Multi-pose Guided Virtual Try-on Network (MG-VTON)\ncan generate a new person image after fitting the desired clothes into the\ninput image and manipulating human poses. Our MG-VTON is constructed in three\nstages: 1) a desired human parsing map of the target image is synthesized to\nmatch both the desired pose and the desired clothes shape; 2) a deep Warping\nGenerative Adversarial Network (Warp-GAN) warps the desired clothes appearance\ninto the synthesized human parsing map and alleviates the misalignment problem\nbetween the input human pose and desired human pose; 3) a refinement render\nutilizing multi-pose composition masks recovers the texture details of clothes\nand removes some artifacts. Extensive experiments on well-known datasets and\nour newly collected largest virtual try-on benchmark demonstrate that our\nMG-VTON significantly outperforms all state-of-the-art methods both\nqualitatively and quantitatively with promising multi-pose virtual try-on\nperformances.","url_abs":"http://arxiv.org/abs/1902.11026v1","url_pdf":"http://arxiv.org/pdf/1902.11026v1.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":[],"tasks":[{"task_slug":"fashion-synthesis","task_name":"Fashion Synthesis"},{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"human-parsing","task_name":"Human Parsing"},{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"},{"task_slug":"virtual-try-on","task_name":"Virtual Try-on"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/virtual-try-on-on-deep-fashion","task":"Virtual Try-on","dataset":"Deep-Fashion","model":"MG-VTON","rank_in_archive_order":1,"of":2,"metrics":{"IS":" 3.03","SSIM":"0.744"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1902.11026","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}