{"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/dmcnn-dual-domain-multi-scale-convolutional","title":"DMCNN: Dual-Domain Multi-Scale Convolutional Neural Network for Compression Artifacts Removal","arxiv_id":"1806.03275","date":"2018-06-08","proceeding":null,"authors":["Xiaoshuai Zhang","Wenhan Yang","Yueyu Hu","Jiaying Liu"],"abstract":"JPEG is one of the most commonly used standards among lossy image compression\nmethods. However, JPEG compression inevitably introduces various kinds of\nartifacts, especially at high compression rates, which could greatly affect the\nQuality of Experience (QoE). Recently, convolutional neural network (CNN) based\nmethods have shown excellent performance for removing the JPEG artifacts. Lots\nof efforts have been made to deepen the CNNs and extract deeper features, while\nrelatively few works pay attention to the receptive field of the network. In\nthis paper, we illustrate that the quality of output images can be\nsignificantly improved by enlarging the receptive fields in many cases. One\nstep further, we propose a Dual-domain Multi-scale CNN (DMCNN) to take full\nadvantage of redundancies on both the pixel and DCT domains. Experiments show\nthat DMCNN sets a new state-of-the-art for the task of JPEG artifact removal.","url_abs":"http://arxiv.org/abs/1806.03275v2","url_pdf":"http://arxiv.org/pdf/1806.03275v2.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":[],"tasks":[{"task_slug":"image-compression","task_name":"Image Compression"},{"task_slug":"jpeg-artifact-correction","task_name":"JPEG Artifact Correction"},{"task_slug":"jpeg-artifact-removal","task_name":"JPEG Artifact Removal"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-10","task":"JPEG Artifact Correction","dataset":"ICB (Quality 10 Color)","model":"DMCNN","rank_in_archive_order":4,"of":6,"metrics":{"PSNR":"30.85","PSNR-B":"31.31","SSIM":"0.796"},"uses_additional_data":false},{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-10-1","task":"JPEG Artifact Correction","dataset":"ICB (Quality 10 Grayscale)","model":"DMCNN","rank_in_archive_order":2,"of":5,"metrics":{"PSNR":"34.18","PSNR-B":"34.15","SSIM":"0.874"},"uses_additional_data":false},{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-20","task":"JPEG Artifact Correction","dataset":"ICB (Quality 20 Color)","model":"DMCNN","rank_in_archive_order":5,"of":6,"metrics":{"PSNR":"32.77","PSNR-B":"33.26","SSIM":"0.830"},"uses_additional_data":false},{"leaderboard":"/sota/jpeg-artifact-correction-on-icb-quality-20-1","task":"JPEG Artifact Correction","dataset":"ICB (Quality 20 Grayscale)","model":"DMCNN","rank_in_archive_order":3,"of":5,"metrics":{"PSNR":"35.93","PSNR-B":"35.79","SSIM":"0.918"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.03275","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}