{"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/improving-shape-deformation-in-unsupervised","title":"Improving Shape Deformation in Unsupervised Image-to-Image Translation","arxiv_id":"1808.04325","date":"2018-08-13","proceeding":"ECCV 2018 9","authors":["Aaron Gokaslan","Vivek Ramanujan","Daniel Ritchie","Kwang In Kim","James Tompkin"],"abstract":"Unsupervised image-to-image translation techniques are able to map local\ntexture between two domains, but they are typically unsuccessful when the\ndomains require larger shape change. Inspired by semantic segmentation, we\nintroduce a discriminator with dilated convolutions that is able to use\ninformation from across the entire image to train a more context-aware\ngenerator. This is coupled with a multi-scale perceptual loss that is better\nable to represent error in the underlying shape of objects. We demonstrate that\nthis design is more capable of representing shape deformation in a challenging\ntoy dataset, plus in complex mappings with significant dataset variation\nbetween humans, dolls, and anime faces, and between cats and dogs.","url_abs":"http://arxiv.org/abs/1808.04325v2","url_pdf":"http://arxiv.org/pdf/1808.04325v2.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":"improving-shape-deformation-in-unsupervised","repo_url":"https://github.com/brownvc/ganimorph","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"improving-shape-deformation-in-unsupervised","repo_url":"https://github.com/Funaizhang/dics","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"improving-shape-deformation-in-unsupervised","repo_url":"https://github.com/advaitrane/GANimorph_pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"improving-shape-deformation-in-unsupervised","repo_url":"https://github.com/itsss/ganimorph","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"unsupervised-image-to-image-translation","task_name":"Unsupervised Image-To-Image Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.04325","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}