{"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/ug2-a-video-benchmark-for-assessing-the","title":"UG^2: a Video Benchmark for Assessing the Impact of Image Restoration and Enhancement on Automatic Visual Recognition","arxiv_id":"1710.02909","date":"2017-10-09","proceeding":null,"authors":["Rosaura G. Vidal","Sreya Banerjee","Klemen Grm","Vitomir Struc","Walter J. Scheirer"],"abstract":"Advances in image restoration and enhancement techniques have led to\ndiscussion about how such algorithmscan be applied as a pre-processing step to\nimprove automatic visual recognition. In principle, techniques like deblurring\nand super-resolution should yield improvements by de-emphasizing noise and\nincreasing signal in an input image. But the historically divergent goals of\nthe computational photography and visual recognition communities have created a\nsignificant need for more work in this direction. To facilitate new research,\nwe introduce a new benchmark dataset called UG^2, which contains three\ndifficult real-world scenarios: uncontrolled videos taken by UAVs and manned\ngliders, as well as controlled videos taken on the ground. Over 160,000\nannotated frames forhundreds of ImageNet classes are available, which are used\nfor baseline experiments that assess the impact of known and unknown image\nartifacts and other conditions on common deep learning-based object\nclassification approaches. Further, current image restoration and enhancement\ntechniques are evaluated by determining whether or not theyimprove baseline\nclassification performance. Results showthat there is plenty of room for\nalgorithmic innovation, making this dataset a useful tool going forward.","url_abs":"http://arxiv.org/abs/1710.02909v2","url_pdf":"http://arxiv.org/pdf/1710.02909v2.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":[],"tasks":[{"task_slug":"deblurring","task_name":"Deblurring"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-restoration","task_name":"Image Restoration"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[],"datasets_introduced":[{"slug":"ug-2","name":"UG^2","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1710.02909","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}