{"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/getting-to-know-low-light-images-with-the","title":"Getting to Know Low-light Images with The Exclusively Dark Dataset","arxiv_id":"1805.11227","date":"2018-05-29","proceeding":null,"authors":["Yuen Peng Loh","Chee Seng Chan"],"abstract":"Low-light is an inescapable element of our daily surroundings that greatly\naffects the efficiency of our vision. Research works on low-light has seen a\nsteady growth, particularly in the field of image enhancement, but there is\nstill a lack of a go-to database as benchmark. Besides, research fields that\nmay assist us in low-light environments, such as object detection, has glossed\nover this aspect even though breakthroughs-after-breakthroughs had been\nachieved in recent years, most noticeably from the lack of low-light data (less\nthan 2% of the total images) in successful public benchmark dataset such as\nPASCAL VOC, ImageNet, and Microsoft COCO. Thus, we propose the Exclusively Dark\ndataset to elevate this data drought, consisting exclusively of ten different\ntypes of low-light images (i.e. low, ambient, object, single, weak, strong,\nscreen, window, shadow and twilight) captured in visible light only with image\nand object level annotations. Moreover, we share insightful findings in regards\nto the effects of low-light on the object detection task by analyzing\nvisualizations of both hand-crafted and learned features. Most importantly, we\nfound that the effects of low-light reaches far deeper into the features than\ncan be solved by simple \"illumination invariance'\". It is our hope that this\nanalysis and the Exclusively Dark dataset can encourage the growth in low-light\ndomain researches on different fields. The Exclusively Dark dataset with its\nannotation is available at\nhttps://github.com/cs-chan/Exclusively-Dark-Image-Dataset","url_abs":"http://arxiv.org/abs/1805.11227v1","url_pdf":"http://arxiv.org/pdf/1805.11227v1.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":"getting-to-know-low-light-images-with-the","repo_url":"https://github.com/cs-chan/Exclusively-Dark-Image-Dataset","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"getting-to-know-low-light-images-with-the","repo_url":"https://github.com/cuiziteng/illumination-adaptive-transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"getting-to-know-low-light-images-with-the","repo_url":"https://github.com/Zeng555/EMV-YOLO","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"image-enhancement","task_name":"Image Enhancement"},{"task_slug":"low-light-image-enhancement","task_name":"Low-Light Image Enhancement"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[{"slug":"exdark","name":"ExDark","full_name":"Exclusively Dark Image Dataset"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.11227","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}