{"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/fast-perceptual-image-enhancement","title":"Fast Perceptual Image Enhancement","arxiv_id":"1812.11852","date":"2018-12-31","proceeding":null,"authors":["Etienne de Stoutz","Andrey Ignatov","Nikolay Kobyshev","Radu Timofte","Luc van Gool"],"abstract":"The vast majority of photos taken today are by mobile phones. While their\nquality is rapidly growing, due to physical limitations and cost constraints,\nmobile phone cameras struggle to compare in quality with DSLR cameras. This\nmotivates us to computationally enhance these images. We extend upon the\nresults of Ignatov et al., where they are able to translate images from compact\nmobile cameras into images with comparable quality to high-resolution photos\ntaken by DSLR cameras. However, the neural models employed require large\namounts of computational resources and are not lightweight enough to run on\nmobile devices. We build upon the prior work and explore different network\narchitectures targeting an increase in image quality and speed. With an\nefficient network architecture which does most of its processing in a lower\nspatial resolution, we achieve a significantly higher mean opinion score (MOS)\nthan the baseline while speeding up the computation by 6.3 times on a\nconsumer-grade CPU. This suggests a promising direction for\nneural-network-based photo enhancement using the phone hardware of the future.","url_abs":"http://arxiv.org/abs/1812.11852v1","url_pdf":"http://arxiv.org/pdf/1812.11852v1.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":"fast-perceptual-image-enhancement","repo_url":"https://github.com/dojure/FPIE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":null,"task_name":"CPU"},{"task_slug":"image-enhancement","task_name":"Image Enhancement"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}