{"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/a-fully-progressive-approach-to-single-image","title":"A Fully Progressive Approach to Single-Image Super-Resolution","arxiv_id":"1804.02900","date":"2018-04-09","proceeding":null,"authors":["Yifan Wang","Federico Perazzi","Brian McWilliams","Alexander Sorkine-Hornung","Olga Sorkine-Hornung","Christopher Schroers"],"abstract":"Recent deep learning approaches to single image super-resolution have\nachieved impressive results in terms of traditional error measures and\nperceptual quality. However, in each case it remains challenging to achieve\nhigh quality results for large upsampling factors. To this end, we propose a\nmethod (ProSR) that is progressive both in architecture and training: the\nnetwork upsamples an image in intermediate steps, while the learning process is\norganized from easy to hard, as is done in curriculum learning. To obtain more\nphotorealistic results, we design a generative adversarial network (GAN), named\nProGanSR, that follows the same progressive multi-scale design principle. This\nnot only allows to scale well to high upsampling factors (e.g., 8x) but\nconstitutes a principled multi-scale approach that increases the reconstruction\nquality for all upsampling factors simultaneously. In particular ProSR ranks\n2nd in terms of SSIM and 4th in terms of PSNR in the NTIRE2018 SISR challenge\n[34]. Compared to the top-ranking team, our model is marginally lower, but runs\n5 times faster.","url_abs":"http://arxiv.org/abs/1804.02900v2","url_pdf":"http://arxiv.org/pdf/1804.02900v2.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":"a-fully-progressive-approach-to-single-image","repo_url":"https://github.com/anktplwl91/Image-Superresolution","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"a-fully-progressive-approach-to-single-image","repo_url":"https://github.com/fperazzi/proSR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"a-fully-progressive-approach-to-single-image","repo_url":"https://github.com/robo-warrior/SR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"a-fully-progressive-approach-to-single-image","repo_url":"https://github.com/singnet/super-resolution-service","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"a-fully-progressive-approach-to-single-image","repo_url":"https://github.com/vskulkarni31/proSR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"a-fully-progressive-approach-to-single-image","repo_url":"https://github.com/vskulkarni31/srgan","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"ssim","task_name":"SSIM"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-super-resolution-on-bsd100-4x-upscaling","task":"Image Super-Resolution","dataset":"BSD100 - 4x upscaling","model":"ProSR","rank_in_archive_order":20,"of":71,"metrics":{"PSNR":"27.79"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set14-4x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 4x upscaling","model":"ProSR","rank_in_archive_order":37,"of":104,"metrics":{"PSNR":"28.94"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-4x","task":"Image Super-Resolution","dataset":"Urban100 - 4x upscaling","model":"ProSR","rank_in_archive_order":26,"of":65,"metrics":{"PSNR":"26.89"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.02900","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}