{"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/learning-to-detect-multiple-photographic","title":"Learning to Detect Multiple Photographic Defects","arxiv_id":"1612.01635","date":"2016-12-06","proceeding":null,"authors":["Ning Yu","Xiaohui Shen","Zhe Lin","Radomir Mech","Connelly Barnes"],"abstract":"In this paper, we introduce the problem of simultaneously detecting multiple\nphotographic defects. We aim at detecting the existence, severity, and\npotential locations of common photographic defects related to color, noise,\nblur and composition. The automatic detection of such defects could be used to\nprovide users with suggestions for how to improve photos without the need to\nlaboriously try various correction methods. Defect detection could also help\nusers select photos of higher quality while filtering out those with severe\ndefects in photo curation and summarization.\n  To investigate this problem, we collected a large-scale dataset of user\nannotations on seven common photographic defects, which allows us to evaluate\nalgorithms by measuring their consistency with human judgments. Our new dataset\nenables us to formulate the problem as a multi-task learning problem and train\na multi-column deep convolutional neural network (CNN) to simultaneously\npredict the severity of all the defects. Unlike some existing single-defect\nestimation methods that rely on low-level statistics and may fail in many cases\non natural photographs, our model is able to understand image contents and\nquality at a higher level. As a result, in our experiments, we show that our\nmodel has predictions with much higher consistency with human judgments than\nlow-level methods as well as several baseline CNN models. Our model also\nperforms better than an average human from our user study.","url_abs":"http://arxiv.org/abs/1612.01635v5","url_pdf":"http://arxiv.org/pdf/1612.01635v5.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":"learning-to-detect-multiple-photographic","repo_url":"https://github.com/ningyu1991/DefectDetection","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"caffe2","reach":null}],"tasks":[{"task_slug":"defect-detection","task_name":"Defect Detection"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"}],"methods":[],"datasets_introduced":[{"slug":"photographic-defect-severity","name":"Photographic Defect Severity","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.01635","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}