{"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/exposure-a-white-box-photo-post-processing","title":"Exposure: A White-Box Photo Post-Processing Framework","arxiv_id":"1709.09602","date":"2017-09-27","proceeding":null,"authors":["Yuanming Hu","Hao He","Chenxi Xu","Baoyuan Wang","Stephen Lin"],"abstract":"Retouching can significantly elevate the visual appeal of photos, but many\ncasual photographers lack the expertise to do this well. To address this\nproblem, previous works have proposed automatic retouching systems based on\nsupervised learning from paired training images acquired before and after\nmanual editing. As it is difficult for users to acquire paired images that\nreflect their retouching preferences, we present in this paper a deep learning\napproach that is instead trained on unpaired data, namely a set of photographs\nthat exhibits a retouching style the user likes, which is much easier to\ncollect. Our system is formulated using deep convolutional neural networks that\nlearn to apply different retouching operations on an input image. Network\ntraining with respect to various types of edits is enabled by modeling these\nretouching operations in a unified manner as resolution-independent\ndifferentiable filters. To apply the filters in a proper sequence and with\nsuitable parameters, we employ a deep reinforcement learning approach that\nlearns to make decisions on what action to take next, given the current state\nof the image. In contrast to many deep learning systems, ours provides users\nwith an understandable solution in the form of conventional retouching edits,\nrather than just a \"black-box\" result. Through quantitative comparisons and\nuser studies, we show that this technique generates retouching results\nconsistent with the provided photo set.","url_abs":"http://arxiv.org/abs/1709.09602v2","url_pdf":"http://arxiv.org/pdf/1709.09602v2.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":"exposure-a-white-box-photo-post-processing","repo_url":"https://github.com/yuanming-hu/exposure","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}