{"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/image-reconstruction-with-predictive-filter","title":"Image Reconstruction with Predictive Filter Flow","arxiv_id":"1811.11482","date":"2018-11-28","proceeding":null,"authors":["Shu Kong","Charless Fowlkes"],"abstract":"We propose a simple, interpretable framework for solving a wide range of\nimage reconstruction problems such as denoising and deconvolution. Given a\ncorrupted input image, the model synthesizes a spatially varying linear filter\nwhich, when applied to the input image, reconstructs the desired output. The\nmodel parameters are learned using supervised or self-supervised training. We\ntest this model on three tasks: non-uniform motion blur removal,\nlossy-compression artifact reduction and single image super resolution. We\ndemonstrate that our model substantially outperforms state-of-the-art methods\non all these tasks and is significantly faster than optimization-based\napproaches to deconvolution. Unlike models that directly predict output pixel\nvalues, the predicted filter flow is controllable and interpretable, which we\ndemonstrate by visualizing the space of predicted filters for different tasks.","url_abs":"http://arxiv.org/abs/1811.11482v1","url_pdf":"http://arxiv.org/pdf/1811.11482v1.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":"image-reconstruction-with-predictive-filter","repo_url":"https://github.com/aimerykong/predictive-filter-flow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"image-reconstruction-with-predictive-filter","repo_url":"https://github.com/bestaar/predictiveFilterFlow","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"deblurring","task_name":"Deblurring"},{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"image-reconstruction","task_name":"Image Reconstruction"},{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"lossy-compression-artifact-reduction","task_name":"Lossy-Compression Artifact Reduction"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-super-resolution-on-set14-4x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 4x upscaling","model":"PFF","rank_in_archive_order":32,"of":104,"metrics":{"PSNR":"28.98","SSIM":"0.7904"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.11482","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}