{"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/on-learning-optimized-reaction-diffusion","title":"On learning optimized reaction diffusion processes for effective image restoration","arxiv_id":"1503.05768","date":"2015-03-19","proceeding":"CVPR 2015 6","authors":["Yunjin Chen","Wei Yu","Thomas Pock"],"abstract":"For several decades, image restoration remains an active research topic in\nlow-level computer vision and hence new approaches are constantly emerging.\nHowever, many recently proposed algorithms achieve state-of-the-art performance\nonly at the expense of very high computation time, which clearly limits their\npractical relevance. In this work, we propose a simple but effective approach\nwith both high computational efficiency and high restoration quality. We extend\nconventional nonlinear reaction diffusion models by several parametrized linear\nfilters as well as several parametrized influence functions. We propose to\ntrain the parameters of the filters and the influence functions through a loss\nbased approach. Experiments show that our trained nonlinear reaction diffusion\nmodels largely benefit from the training of the parameters and finally lead to\nthe best reported performance on common test datasets for image restoration.\nDue to their structural simplicity, our trained models are highly efficient and\nare also well-suited for parallel computation on GPUs.","url_abs":"http://arxiv.org/abs/1503.05768v2","url_pdf":"http://arxiv.org/pdf/1503.05768v2.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":"on-learning-optimized-reaction-diffusion","repo_url":"https://github.com/VLOGroup/mri-variationalnetwork","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"on-learning-optimized-reaction-diffusion","repo_url":"https://github.com/VLOGroup/tensorflow-icg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"on-learning-optimized-reaction-diffusion","repo_url":"https://github.com/jplumail/learning-image-restoration","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":"image-restoration","task_name":"Image Restoration"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1503.05768","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}