{"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/unsupervised-image-super-resolution-using","title":"Unsupervised Image Super-Resolution using Cycle-in-Cycle Generative Adversarial Networks","arxiv_id":"1809.00437","date":"2018-09-03","proceeding":null,"authors":["Yuan Yuan","Siyuan Liu","Jiawei Zhang","Yongbing Zhang","Chao Dong","Liang Lin"],"abstract":"We consider the single image super-resolution problem in a more general case\nthat the low-/high-resolution pairs and the down-sampling process are\nunavailable. Different from traditional super-resolution formulation, the\nlow-resolution input is further degraded by noises and blurring. This\ncomplicated setting makes supervised learning and accurate kernel estimation\nimpossible. To solve this problem, we resort to unsupervised learning without\npaired data, inspired by the recent successful image-to-image translation\napplications. With generative adversarial networks (GAN) as the basic\ncomponent, we propose a Cycle-in-Cycle network structure to tackle the problem\nwithin three steps. First, the noisy and blurry input is mapped to a noise-free\nlow-resolution space. Then the intermediate image is up-sampled with a\npre-trained deep model. Finally, we fine-tune the two modules in an end-to-end\nmanner to get the high-resolution output. Experiments on NTIRE2018 datasets\ndemonstrate that the proposed unsupervised method achieves comparable results\nas the state-of-the-art supervised models.","url_abs":"http://arxiv.org/abs/1809.00437v1","url_pdf":"http://arxiv.org/pdf/1809.00437v1.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":"unsupervised-image-super-resolution-using","repo_url":"https://github.com/sangyun884/CinCGAN-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"},{"task_slug":"super-resolution","task_name":"Super-Resolution"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.00437","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}