{"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-generic-deep-architecture-for-single-image","title":"A Generic Deep Architecture for Single Image Reflection Removal and Image Smoothing","arxiv_id":"1708.03474","date":"2017-08-11","proceeding":"ICCV 2017 10","authors":["Qingnan Fan","Jiaolong Yang","Gang Hua","Baoquan Chen","David Wipf"],"abstract":"This paper proposes a deep neural network structure that exploits edge\ninformation in addressing representative low-level vision tasks such as layer\nseparation and image filtering. Unlike most other deep learning strategies\napplied in this context, our approach tackles these challenging problems by\nestimating edges and reconstructing images using only cascaded convolutional\nlayers arranged such that no handcrafted or application-specific\nimage-processing components are required. We apply the resulting transferrable\npipeline to two different problem domains that are both sensitive to edges,\nnamely, single image reflection removal and image smoothing. For the former,\nusing a mild reflection smoothness assumption and a novel synthetic data\ngeneration method that acts as a type of weak supervision, our network is able\nto solve much more difficult reflection cases that cannot be handled by\nprevious methods. For the latter, we also exceed the state-of-the-art\nquantitative and qualitative results by wide margins. In all cases, the\nproposed framework is simple, fast, and easy to transfer across disparate\ndomains.","url_abs":"http://arxiv.org/abs/1708.03474v2","url_pdf":"http://arxiv.org/pdf/1708.03474v2.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-generic-deep-architecture-for-single-image","repo_url":"https://github.com/fqnchina/CEILNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"torch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"reflection-removal","task_name":"Reflection Removal"},{"task_slug":"synthetic-data-generation","task_name":"Synthetic Data Generation"},{"task_slug":"image-smoothing","task_name":"image smoothing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1708.03474","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}