{"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/image-quality-assessment-guided-deep-neural","title":"Image Quality Assessment Guided Deep Neural Networks Training","arxiv_id":"1708.03880","date":"2017-08-13","proceeding":null,"authors":["Zhuo Chen","Weisi Lin","Shiqi Wang","Long Xu","Leida Li"],"abstract":"For many computer vision problems, the deep neural networks are trained and\nvalidated based on the assumption that the input images are pristine (i.e.,\nartifact-free). However, digital images are subject to a wide range of\ndistortions in real application scenarios, while the practical issues regarding\nimage quality in high level visual information understanding have been largely\nignored. In this paper, in view of the fact that most widely deployed deep\nlearning models are susceptible to various image distortions, the distorted\nimages are involved for data augmentation in the deep neural network training\nprocess to learn a reliable model for practical applications. In particular, an\nimage quality assessment based label smoothing method, which aims at\nregularizing the label distribution of training images, is further proposed to\ntune the objective functions in learning the neural network. Experimental\nresults show that the proposed method is effective in dealing with both low and\nhigh quality images in the typical image classification task.","url_abs":"http://arxiv.org/abs/1708.03880v1","url_pdf":"http://arxiv.org/pdf/1708.03880v1.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":"image-quality-assessment-guided-deep-neural","repo_url":"https://github.com/dzuba29/Deeplom","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"image-quality-assessment-guided-deep-neural","repo_url":"https://github.com/dzuba29/Image-quality-assesment","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"image-quality-assessment-guided-deep-neural","repo_url":"https://github.com/dzuba29/Image-quality-assessment","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-quality-assessment","task_name":"Image Quality Assessment"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"label-smoothing","method_name":"Label Smoothing"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}