{"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/deep-high-dynamic-range-imaging-with-large","title":"Deep High Dynamic Range Imaging with Large Foreground Motions","arxiv_id":"1711.08937","date":"2017-11-24","proceeding":"ECCV 2018 9","authors":["Shangzhe Wu","Jiarui Xu","Yu-Wing Tai","Chi-Keung Tang"],"abstract":"This paper proposes the first non-flow-based deep framework for high dynamic\nrange (HDR) imaging of dynamic scenes with large-scale foreground motions. In\nstate-of-the-art deep HDR imaging, input images are first aligned using optical\nflows before merging, which are still error-prone due to occlusion and large\nmotions. In stark contrast to flow-based methods, we formulate HDR imaging as\nan image translation problem without optical flows. Moreover, our simple\ntranslation network can automatically hallucinate plausible HDR details in the\npresence of total occlusion, saturation and under-exposure, which are otherwise\nalmost impossible to recover by conventional optimization approaches. Our\nframework can also be extended for different reference images. We performed\nextensive qualitative and quantitative comparisons to show that our approach\nproduces excellent results where color artifacts and geometric distortions are\nsignificantly reduced compared to existing state-of-the-art methods, and is\nrobust across various inputs, including images without radiometric calibration.","url_abs":"http://arxiv.org/abs/1711.08937v3","url_pdf":"http://arxiv.org/pdf/1711.08937v3.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":"deep-high-dynamic-range-imaging-with-large","repo_url":"https://github.com/elliottwu/DeepHDR","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"translation","task_name":"Translation"},{"task_slug":"high","task_name":"Vocal Bursts Intensity Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.08937","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}