[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-512781-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-512781-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","toward-convolutional-blind-denoising-of-real-photographs","Toward Convolutional Blind Denoising of Real Photographs","","Deep CNNs have advanced image denoising under additive white Gaussian noise (AWGN), yet their results on real photographs stay limited because the training objective overfits a simplified noise distribution that diverges from complex camera noise. CBDNet trains a convolutional blind denoising network using a realistic noise model plus paired real noisy and nearly noise-free images. It models Poisson-Gaussian signal-dependent noise and accounts for in-camera processing, and embeds a noise-estimation subnetwork with asymmetric learning. Experiments on real noisy datasets show clear gains in quantitative metrics and visual quality, with available public code.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/toward-convolutional-blind-denoising-of-real-photographs/512781/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/toward-convolutional-blind-denoising-of-real-photographs/512781.png","ImageObject",300,407,{"name":42,"@type":43},"วิน","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-13","2026-10-06",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",16,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why do deep CNN denoisers trained on AWGN perform poorly on real noisy photographs?","Question",{"text":63,"@type":64},"Their learned models overfit the simplified AWGN noise distribution, which deviates from real-world camera noise characteristics and the effects of in-camera processing.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What does CBDNet change to improve generalization for blind denoising?",{"text":68,"@type":64},"It trains a convolutional blind denoising network with more realistic noise modeling (Poisson-Gaussian plus in-camera processing effects) and includes both synthetic and real noisy-clean image pairs.",{"name":70,"@type":61,"acceptedAnswer":71},"How does CBDNet provide interactive improvement of denoising results?",{"text":72,"@type":64},"CBDNet includes a noise estimation subnetwork with asymmetric learning to suppress under-estimation, and it allows users to adjust the estimated noise level map to rectify the denoising output.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},512781,1791376271,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":140,"language":141,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":12,"update_tm":145,"read_time":146,"preview_page_count":33,"preview_urls":147},2336475104736,"https://ap-avatar.wpscdn.com/avatar/22000c4c5e0e5b17e70?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786591360781797222","Toward Convolutional Blind Denoising of Real Photographs  \nShi Guo 1 ;3 ;4 , Zifei Yan (☞) 1 , Kai Zhang 1 ;3 , Wangmeng Zuo 1 ;2 , Lei Zhang3 ;4  \n1Harbin Institute of Technology, Harbin; 2Peng Cheng Laboratory, Shenzhen;  \n3 The Hong Kong Polytechnic University, Hong Kong; 4DAMO Academy, Alibaba Group  \n[guoshi28@outlook.com](guoshi28@outlook.com) , fwmzuo,[yanzifei](yanzifeig@hit.edu.cn)[g](yanzifeig@hit.edu.cn)[@hit.edu.cn](yanzifeig@hit.edu.cn)  \n[cskaizhang@gmail.com](cskaizhang@gmail.com) , [cslzhang@comp.polyu.edu.hk](cslzhang@comp.polyu.edu.hk)  \nAbstract  \nWhile deep convolutional neural networks (CNNs) have achieved impressive success in image denoising with additive white Gaussian noise (AWGN), their performance remains limited on real-world noisy photographs. The main reason is that their learned models are easy to overﬁt on the simpliﬁed AWGN model which deviates severely from the complicated real-world noise model. In order to improve the generalization ability of deep CNN denoisers, we suggest training a convolutional blind denoising network (CBDNet) with more realistic noise model and real-world noisy-clean image pairs. On the one hand, both signaldependent noise and in-camera signal processing pipeline is considered to synthesize realistic noisy images. On the other hand, real-world noisy photographs and their nearly noise-free counterparts are also included to train our CBDNet. To further provide an interactive strategy to rectify denoising result conveniently, a noise estimation subnetwork with asymmetric learning to suppress under-estimation of noise level is embedded into CBDNet. Extensive experimental results on three datasets of real-world noisy photographs clearly demonstrate the superior performance of CBDNet over state-of-the-arts in terms of quantitative metrics and visual quality. The code has been made available at [https://github.com/GuoShi28/CBDNet](https://github.com/GuoShi28/CBDNet).  \n1. Introduction  \nImage denoising is an essential and fundamental problem in low-level vision and image processing. With decades of studies, numerous promising approaches [3, 12, 17, 53, 11, 61] have been developed and near-optimal performance [8, 31, 50] has been achieved for the removal of additive white Gaussian noise (AWGN) . However, in real camera system, image noise comes from multiple sources (e.g., dark current noise, short noise, and thermal noise) and is further affected by in-camera processing (ISP) pipeline (e.g., demosaicing, Gamma correction, and compression) . All these make real noise much more different  \n(a)“0002 02” from DND [45]  \n(d) DnCNN [61]  \n(e) FFDNet+ [62]  \nFigure 1: Denoising results of different methods on realworld noisy image “0002 02” from DND [45] .  \nfrom AWGN, and blind denoising of real-world noisy photographs remains a challenging issue.  \nIn the recent past, Gaussian denoising performance has been signiﬁcantly advanced by the development of deep CNNs [61, 38, 62] . However, deep denoisers for blind AWGN removal degrades dramatically when applied to real photographs (see Fig. 1(d)) . On the other hand, deep denoisers for non-blind AWGN removal would smooth out the details while removing the noise (see Fig. 1(e)) . Such an phenomenon may be explained from the characteristic of deep CNNs [39], where their generalization largely depends on the ability of memorizing large scale training data. In other words, existing CNN denoisers tend to be over-ﬁtted to Gaussian noise and generalize poorly to real-world noisy images with more sophisticated noise.  \nIn this paper, we tackle this issue by developing a convolutional blind denoising network (CBDNet) for real-world photographs. As indicated by [39], the success of CNN denoisers are signiﬁcantly dependent on whether the distributions of synthetic and real noises are well matched. Therefore, realistic noise model is the foremost issue for blind denoising of real photographs. According to [14, 45], PoissonGaussian distribution which","cbCaidbTFNUHb7RT","https://ap.wps.com/l/cbCaidbTFNUHb7RT","pdf",5197546,11,"English","# Abstract\n# Introduction","[{\"question\":\"Why do deep CNN denoisers trained on AWGN perform poorly on real noisy photographs?\",\"answer\":\"Their learned models overfit the simplified AWGN noise distribution, which deviates from real-world camera noise characteristics and the effects of in-camera processing.\"},{\"question\":\"What does CBDNet change to improve generalization for blind denoising?\",\"answer\":\"It trains a convolutional blind denoising network with more realistic noise modeling (Poisson-Gaussian plus in-camera processing effects) and includes both synthetic and real noisy-clean image pairs.\"},{\"question\":\"How does CBDNet provide interactive improvement of denoising results?\",\"answer\":\"CBDNet includes a noise estimation subnetwork with asymmetric learning to suppress under-estimation, and it allows users to adjust the estimated noise level map to rectify the denoising output.\"}]","Toward Convolutional Blind Denoising of Real Photographs | PDF",1791275480,28,[148,152,156,160],{"page_no":22,"url":149,"width":150,"height":151},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/08e1e35521ae7c633ad36c153cf9aae5.webp",983,1272,{"page_no":26,"url":153,"width":154,"height":155},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/63c143b8952677041d65d4ff999c91f2.webp",982,1270,{"page_no":30,"url":157,"width":158,"height":159},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/3a2fd734df15ec6c3b9d2c6cae5316d5.webp",981,1269,{"page_no":33,"url":161,"width":162,"height":163},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/27e67c36d1c6d0016c6a4c090827d85e.webp",980,1268]