{"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/dpw-sdnet-dual-pixel-wavelet-domain-deep-cnns","title":"DPW-SDNet: Dual Pixel-Wavelet Domain Deep CNNs for Soft Decoding of JPEG-Compressed Images","arxiv_id":"1805.10558","date":"2018-05-27","proceeding":null,"authors":["Honggang Chen","Xiaohai He","Linbo Qing","Shuhua Xiong","Truong Q. Nguyen"],"abstract":"JPEG is one of the widely used lossy compression methods. JPEG-compressed\nimages usually suffer from compression artifacts including blocking and\nblurring, especially at low bit-rates. Soft decoding is an effective solution\nto improve the quality of compressed images without changing codec or\nintroducing extra coding bits. Inspired by the excellent performance of the\ndeep convolutional neural networks (CNNs) on both low-level and high-level\ncomputer vision problems, we develop a dual pixel-wavelet domain deep\nCNNs-based soft decoding network for JPEG-compressed images, namely DPW-SDNet.\nThe pixel domain deep network takes the four downsampled versions of the\ncompressed image to form a 4-channel input and outputs a pixel domain\nprediction, while the wavelet domain deep network uses the 1-level discrete\nwavelet transformation (DWT) coefficients to form a 4-channel input to produce\na DWT domain prediction. The pixel domain and wavelet domain estimates are\ncombined to generate the final soft decoded result. Experimental results\ndemonstrate the superiority of the proposed DPW-SDNet over several\nstate-of-the-art compression artifacts reduction algorithms.","url_abs":"http://arxiv.org/abs/1805.10558v1","url_pdf":"http://arxiv.org/pdf/1805.10558v1.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":"blocking","task_name":"Blocking"},{"task_slug":"jpeg-artifact-correction","task_name":"JPEG Artifact Correction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/jpeg-artifact-correction-on-live1-quality-10","task":"JPEG Artifact Correction","dataset":"LIVE1 (Quality 10 Color)","model":"DPW-SDNet","rank_in_archive_order":8,"of":9,"metrics":{"PSNR":"27.26","PSNR-B":"27.28","SSIM":"0.803"},"uses_additional_data":false},{"leaderboard":"/sota/jpeg-artifact-correction-on-live1-quality-20","task":"JPEG Artifact Correction","dataset":"LIVE1 (Quality 20 Color)","model":"DPW-SDNet","rank_in_archive_order":8,"of":9,"metrics":{"PSNR":"29.59","PSNR-B":"29.55","SSIM":"0.874"},"uses_additional_data":false},{"leaderboard":"/sota/jpeg-artifact-correction-on-live1-quality-20-1","task":"JPEG Artifact Correction","dataset":"LIVE1 (Quality 20 Grayscale)","model":"DPW-SDNet","rank_in_archive_order":10,"of":12,"metrics":{"PSNR":"31.69","PSNR-B":"31.60","SSIM":"0.8891"},"uses_additional_data":false},{"leaderboard":"/sota/jpeg-artifact-correction-on-live1-quality-10-1","task":"JPEG Artifact Correction","dataset":"Live1 (Quality 10 Grayscale)","model":"DPW-SDNet","rank_in_archive_order":9,"of":13,"metrics":{"PSNR":"29.40","PSNR-B":"29.34","SSIM":"0.8235"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.10558","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}