{"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/heavy-rain-image-restoration-integrating","title":"Heavy Rain Image Restoration: Integrating Physics Model and Conditional Adversarial Learning","arxiv_id":"1904.05050","date":"2019-04-10","proceeding":null,"authors":["Ruotent Li","Loong Fah Cheong","Robby T. Tan"],"abstract":"Most deraining works focus on rain streaks removal but they cannot deal\nadequately with heavy rain images. In heavy rain, streaks are strongly visible,\ndense rain accumulation or rain veiling effect significantly washes out the\nimage, further scenes are relatively more blurry, etc. In this paper, we\npropose a novel method to address these problems. We put forth a 2-stage\nnetwork: a physics-based backbone followed by a depth-guided GAN refinement.\nThe first stage estimates the rain streaks, the transmission, and the\natmospheric light governed by the underlying physics. To tease out these\ncomponents more reliably, a guided filtering framework is used to decompose the\nimage into its low- and high-frequency components. This filtering is guided by\na rain-free residue image --- its content is used to set the passbands for the\ntwo channels in a spatially-variant manner so that the background details do\nnot get mixed up with the rain-streaks. For the second stage, the refinement\nstage, we put forth a depth-guided GAN to recover the background details failed\nto be retrieved by the first stage, as well as correcting artefacts introduced\nby that stage. We have evaluated our method against the state of the art\nmethods. Extensive experiments show that our method outperforms them on real\nrain image data, recovering visually clean images with good details.","url_abs":"http://arxiv.org/abs/1904.05050v1","url_pdf":"http://arxiv.org/pdf/1904.05050v1.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":"heavy-rain-image-restoration-integrating","repo_url":"https://github.com/liruoteng/HeavyRainRemoval","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-restoration","task_name":"Image Restoration"},{"task_slug":"rain-removal","task_name":"Rain Removal"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.05050","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}