{"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/channel-partitioned-windowed-attention-and","title":"Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution","arxiv_id":"2407.16232","date":"2024-07-23","proceeding":null,"authors":["Dinh Phu Tran","Dao Duy Hung","Daeyoung Kim"],"abstract":"Recently, window-based attention methods have shown great potential for computer vision tasks, particularly in Single Image Super-Resolution (SISR). However, it may fall short in capturing long-range dependencies and relationships between distant tokens. Additionally, we find that learning on spatial domain does not convey the frequency content of the image, which is a crucial aspect in SISR. To tackle these issues, we propose a new Channel-Partitioned Attention Transformer (CPAT) to better capture long-range dependencies by sequentially expanding windows along the height and width of feature maps. In addition, we propose a novel Spatial-Frequency Interaction Module (SFIM), which incorporates information from spatial and frequency domains to provide a more comprehensive information from feature maps. This includes information about the frequency content and enhances the receptive field across the entire image. Experimental findings show the effectiveness of our proposed modules and architecture. In particular, CPAT surpasses current state-of-the-art methods by up to 0.31dB at x2 SR on Urban100.","url_abs":"https://arxiv.org/abs/2407.16232v2","url_pdf":"https://arxiv.org/pdf/2407.16232v2.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-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-super-resolution-on-bsd100-2x-upscaling","task":"Image Super-Resolution","dataset":"BSD100 - 2x upscaling","model":"CPAT+","rank_in_archive_order":10,"of":30,"metrics":{"PSNR":"32.66","SSIM":"0.9058"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-bsd100-2x-upscaling","task":"Image Super-Resolution","dataset":"BSD100 - 2x upscaling","model":"CPAT","rank_in_archive_order":11,"of":30,"metrics":{"PSNR":"32.64 ","SSIM":"0.9056"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-bsd100-3x-upscaling","task":"Image Super-Resolution","dataset":"BSD100 - 3x upscaling","model":"CPAT+","rank_in_archive_order":6,"of":21,"metrics":{"PSNR":"29.59","SSIM":"0.8177"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-bsd100-3x-upscaling","task":"Image Super-Resolution","dataset":"BSD100 - 3x upscaling","model":"CPAT","rank_in_archive_order":7,"of":21,"metrics":{"PSNR":"29.56","SSIM":"0.8174"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-bsd100-4x-upscaling","task":"Image Super-Resolution","dataset":"BSD100 - 4x upscaling","model":"CPAT+","rank_in_archive_order":7,"of":71,"metrics":{"PSNR":"28.06","SSIM":"0.7532"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-bsd100-4x-upscaling","task":"Image Super-Resolution","dataset":"BSD100 - 4x upscaling","model":"CPAT","rank_in_archive_order":9,"of":71,"metrics":{"PSNR":"28.04","SSIM":"0.7527"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-manga109-2x","task":"Image Super-Resolution","dataset":"Manga109 - 2x upscaling","model":"CPAT+","rank_in_archive_order":8,"of":21,"metrics":{"PSNR":"40.59","SSIM":"0.9816"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-manga109-2x","task":"Image Super-Resolution","dataset":"Manga109 - 2x upscaling","model":"CPAT","rank_in_archive_order":9,"of":21,"metrics":{"PSNR":"40.48","SSIM":"0.9814"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-manga109-3x","task":"Image Super-Resolution","dataset":"Manga109 - 3x upscaling","model":"CPAT+","rank_in_archive_order":7,"of":17,"metrics":{"PSNR":"35.77","SSIM":"0.9563"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-manga109-3x","task":"Image Super-Resolution","dataset":"Manga109 - 3x upscaling","model":"CPAT","rank_in_archive_order":8,"of":17,"metrics":{"PSNR":"35.66","SSIM":"0.9559"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-manga109-4x","task":"Image Super-Resolution","dataset":"Manga109 - 4x upscaling","model":"CPAT+","rank_in_archive_order":7,"of":50,"metrics":{"PSNR":"32.85","SSIM":"0.9318"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set14-2x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 2x upscaling","model":"CPAT+","rank_in_archive_order":7,"of":35,"metrics":{"PSNR":"34.97","SSIM":"0.9280"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set14-2x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 2x upscaling","model":"CPAT","rank_in_archive_order":10,"of":35,"metrics":{"PSNR":"34.91","SSIM":"0.9277"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set14-3x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 3x upscaling","model":"CPAT+","rank_in_archive_order":7,"of":24,"metrics":{"PSNR":"31.19","SSIM":"0.8559"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set14-3x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 3x upscaling","model":"CPAT","rank_in_archive_order":8,"of":24,"metrics":{"PSNR":"31.15","SSIM":"0.8557"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set14-4x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 4x upscaling","model":"CPAT+","rank_in_archive_order":8,"of":104,"metrics":{"PSNR":"29.36","SSIM":"0.7996"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set14-4x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 4x upscaling","model":"CPAT","rank_in_archive_order":10,"of":104,"metrics":{"PSNR":"29.34","SSIM":"0.7991"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set5-2x-upscaling","task":"Image Super-Resolution","dataset":"Set5 - 2x upscaling","model":"CPAT+","rank_in_archive_order":8,"of":41,"metrics":{"PSNR":"38.72","SSIM":"0.9635"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set5-2x-upscaling","task":"Image Super-Resolution","dataset":"Set5 - 2x upscaling","model":"CPAT","rank_in_archive_order":9,"of":41,"metrics":{"PSNR":"38.68","SSIM":"0.9633"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set5-3x-upscaling","task":"Image Super-Resolution","dataset":"Set5 - 3x upscaling","model":"CPAT+","rank_in_archive_order":6,"of":32,"metrics":{"PSNR":"35.19","SSIM":"0.9335"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set5-3x-upscaling","task":"Image Super-Resolution","dataset":"Set5 - 3x upscaling","model":"CPAT","rank_in_archive_order":9,"of":32,"metrics":{"PSNR":"35.16","SSIM":"0.9334"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set5-4x-upscaling","task":"Image Super-Resolution","dataset":"Set5 - 4x upscaling","model":"CPAT+","rank_in_archive_order":3,"of":12,"metrics":{"PSNR":"33.24","SSIM":"0.9071"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set5-4x-upscaling","task":"Image Super-Resolution","dataset":"Set5 - 4x upscaling","model":"CPAT","rank_in_archive_order":5,"of":12,"metrics":{"PSNR":"33.19","SSIM":"0.9069"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-2x","task":"Image Super-Resolution","dataset":"Urban100 - 2x upscaling","model":"CPAT+","rank_in_archive_order":6,"of":29,"metrics":{"PSNR":"34.89","SSIM":"0.9487"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-2x","task":"Image Super-Resolution","dataset":"Urban100 - 2x upscaling","model":"CPAT","rank_in_archive_order":8,"of":29,"metrics":{"PSNR":"34.76","SSIM":"0.9481"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-3x","task":"Image Super-Resolution","dataset":"Urban100 - 3x upscaling","model":"CPAT+","rank_in_archive_order":6,"of":22,"metrics":{"PSNR":"30.63","SSIM":"0.8934"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-3x","task":"Image Super-Resolution","dataset":"Urban100 - 3x upscaling","model":"CPAT","rank_in_archive_order":7,"of":22,"metrics":{"PSNR":"30.52","SSIM":"0.8923"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-4x","task":"Image Super-Resolution","dataset":"Urban100 - 4x upscaling","model":"CPAT+","rank_in_archive_order":8,"of":65,"metrics":{"PSNR":"28.33","SSIM":"0.8425"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-urban100-4x","task":"Image Super-Resolution","dataset":"Urban100 - 4x upscaling","model":"CPAT","rank_in_archive_order":9,"of":65,"metrics":{"PSNR":"28.22","SSIM":"0.8408"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}