{"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/dark-and-bright-channel-prior-embedded","title":"Dark and Bright Channel Prior Embedded Network for Dynamic Scene Deblurring","arxiv_id":null,"date":"2020-05-21","proceeding":null,"authors":["Jianrui Cai","WangMeng Zuo","and Lei Zhang"],"abstract":"Recent years have witnessed the significant progress\r\non convolutional neural networks (CNNs) in dynamic scene\r\ndeblurring. While most of the CNN models are generally learned\r\nby the reconstruction loss defined on training data, incorporating\r\nsuitable image priors as well as regularization terms into the\r\nnetwork architecture could boost the deblurring performance.\r\nIn this work, we propose a Dark and Bright Channel Priors\r\nembedded Network (DBCPeNet) to plug the channel priors into\r\na neural network for effective dynamic scene deblurring. A\r\nnovel trainable dark and bright channel priors embedded layer\r\n(DBCPeL) is developed to aggregate both channel priors and\r\nblurry image representations, and a sparse regularization is\r\nintroduced to regularize the DBCPeNet model learning. Furthermore, we present an effective multi-scale network architecture,\r\nnamely image full scale exploitation (IFSE), which works in both\r\ncoarse-to-fine and fine-to-coarse manners for better exploiting\r\ninformation flow across scales. Experimental results on the GoPro\r\nand Kohler datasets show that our proposed DBCPeNet performs ¨\r\nfavorably against state-of-the-art deep image deblurring methods\r\nin terms of both quantitative metrics and visual quality.","url_abs":"https://ieeexplore.ieee.org/document/9097949","url_pdf":"https://www4.comp.polyu.edu.hk/~cslzhang/paper/DBCPeNet_TIP.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":"dark-and-bright-channel-prior-embedded","repo_url":"https://github.com/csjcai/DBCPeNet","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"deblurring","task_name":"Deblurring"},{"task_slug":"image-deblurring","task_name":"Image Deblurring"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/deblurring-on-gopro","task":"Deblurring","dataset":"GoPro","model":"DBCPeNet","rank_in_archive_order":46,"of":56,"metrics":{"PSNR":"31.10","SSIM":"0.945"},"uses_additional_data":true},{"leaderboard":"/sota/image-deblurring-on-gopro","task":"Image Deblurring","dataset":"GoPro","model":"DBCPeNet","rank_in_archive_order":43,"of":55,"metrics":{"PSNR":"31.10","SSIM":"0.945"},"uses_additional_data":true}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}