{"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/deformable-gans-for-pose-based-human-image","title":"Deformable GANs for Pose-based Human Image Generation","arxiv_id":"1801.00055","date":"2017-12-29","proceeding":"CVPR 2018 6","authors":["Aliaksandr Siarohin","Enver Sangineto","Stephane Lathuiliere","Nicu Sebe"],"abstract":"In this paper we address the problem of generating person images conditioned\non a given pose. Specifically, given an image of a person and a target pose, we\nsynthesize a new image of that person in the novel pose. In order to deal with\npixel-to-pixel misalignments caused by the pose differences, we introduce\ndeformable skip connections in the generator of our Generative Adversarial\nNetwork. Moreover, a nearest-neighbour loss is proposed instead of the common\nL1 and L2 losses in order to match the details of the generated image with the\ntarget image. We test our approach using photos of persons in different poses\nand we compare our method with previous work in this area showing\nstate-of-the-art results in two benchmarks. Our method can be applied to the\nwider field of deformable object generation, provided that the pose of the\narticulated object can be extracted using a keypoint detector.","url_abs":"http://arxiv.org/abs/1801.00055v2","url_pdf":"http://arxiv.org/pdf/1801.00055v2.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":"deformable-gans-for-pose-based-human-image","repo_url":"https://github.com/AliaksandrSiarohin/pose-gan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"gesture-to-gesture-translation","task_name":"Gesture-to-Gesture Translation"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"},{"task_slug":"pose-transfer","task_name":"Pose Transfer"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/gesture-to-gesture-translation-on-ntu-hand","task":"Gesture-to-Gesture Translation","dataset":"NTU Hand Digit","model":"PoseGAN","rank_in_archive_order":5,"of":6,"metrics":{"AMT":"9.3","IS":"2.4017","PSNR":"29.5471"},"uses_additional_data":false},{"leaderboard":"/sota/gesture-to-gesture-translation-on-senz3d","task":"Gesture-to-Gesture Translation","dataset":"Senz3D","model":"PoseGAN","rank_in_archive_order":5,"of":6,"metrics":{"AMT":"8.6","IS":"3.2147","PSNR":"27.3014"},"uses_additional_data":false},{"leaderboard":"/sota/pose-transfer-on-deep-fashion","task":"Pose Transfer","dataset":"Deep-Fashion","model":"Deformable GAN","rank_in_archive_order":9,"of":12,"metrics":{"IS":"3.439","LPIPS":"0.233","Retrieval Top10 Recall":"30.07","SSIM":"0.756"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1801.00055","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}