{"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/powers-of-layers-for-image-to-image","title":"Powers of layers for image-to-image translation","arxiv_id":"2008.05763","date":"2020-08-13","proceeding":null,"authors":["Hugo Touvron","Matthijs Douze","Matthieu Cord","Hervé Jégou"],"abstract":"We propose a simple architecture to address unpaired image-to-image translation tasks: style or class transfer, denoising, deblurring, deblocking, etc. We start from an image autoencoder architecture with fixed weights. For each task we learn a residual block operating in the latent space, which is iteratively called until the target domain is reached. A specific training schedule is required to alleviate the exponentiation effect of the iterations. At test time, it offers several advantages: the number of weight parameters is limited and the compositional design allows one to modulate the strength of the transformation with the number of iterations. This is useful, for instance, when the type or amount of noise to suppress is not known in advance. Experimentally, we provide proofs of concepts showing the interest of our method for many transformations. The performance of our model is comparable or better than CycleGAN with significantly fewer parameters.","url_abs":"https://arxiv.org/abs/2008.05763v1","url_pdf":"https://arxiv.org/pdf/2008.05763v1.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":"deblurring","task_name":"Deblurring"},{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"cycle-consistency-loss","method_name":"Cycle Consistency Loss"},{"method_slug":"gan-least-squares-loss","method_name":"GAN Least Squares Loss"},{"method_slug":"instance-normalization","method_name":"Instance Normalization"},{"method_slug":"patchgan","method_name":"PatchGAN"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-to-image-translation-on-horse2zebra","task":"Image-to-Image Translation","dataset":"horse2zebra","model":"PoL (CycleGAN)","rank_in_archive_order":3,"of":4,"metrics":{"Frechet Inception Distance":"53.0","Number of params":"15.9M"},"uses_additional_data":false},{"leaderboard":"/sota/image-to-image-translation-on-photo2vangogh","task":"Image-to-Image Translation","dataset":"photo2vangogh","model":"PoL (CycleGAN)","rank_in_archive_order":3,"of":3,"metrics":{"Frechet Inception Distance":"152.7","Number of params":"15.9M"},"uses_additional_data":false},{"leaderboard":"/sota/image-to-image-translation-on-vangogh2photo","task":"Image-to-Image Translation","dataset":"vangogh2photo","model":"PoL (CycleGAN)","rank_in_archive_order":2,"of":3,"metrics":{"Frechet Inception Distance":"134.4","Number of Params":"15.9M"},"uses_additional_data":false},{"leaderboard":"/sota/image-to-image-translation-on-zebra2horse","task":"Image-to-Image Translation","dataset":"zebra2horse","model":"PoL (CycleGAN)","rank_in_archive_order":3,"of":3,"metrics":{"Frechet Inception Distance":"112.3","Number of params":"15.9M"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}