摘要
Most deep learning approaches for image quality assessment use regression from deep features extracted by CNN (Convolutional Neural Networks). However, non-local information is usually neglected in exist-ing methods. Motivated by the recent success of transformers in modeling contextual information, we propose a hybrid framework that utilizes a vision transformer backbone to extract features and a CNN decoder for quality estimation. We propose a shared feature extraction scheme for both FR and NR set-tings. A two-branch structured attentive quality predictor is devised for quality prediction. Evaluation experiments on various IQA datasets, including LIVE, CSIQ and TID2013, LIVE-Challenge, KADID-10 K, and KONIQ-10 K, show that our proposed models achieve outstanding performance for both FR and NR settings.& COPY; 2023 Published by Elsevier B.V.
学科与关键词
- Perception-Oriented U-Shaped Transformer Network for 360-Degree No-Reference Image Quality AssessmentZhou, Mingliang; Chen, Lei; Wei, Xuekai*; Liao, Xingran; Mao, Qin*; Wang, Heqiang; Pu, Huayan; Luo, Jun; Xiang, Tao; Fang, BinIEEE Transactions on Broadcasting, 2023, 69(2): 396-405.被引 8
- No-reference image quality assessment by using convolutional neural networks via object detectionCao, Jingchao; Wu, Wenhui; Wang, Ran*; Kwong, SamInternational Journal of Machine Learning and Cybernetics, 2022, 13(11): 3543-3554.被引 2
- An End-to-End Blind Image Quality Assessment Method Using a Recurrent Network and Self-AttentionZhou, Mingliang; Lan, Xuting; Wei, Xuekai*; Liao, Xingran; Mao, Qin*; Li, Yutong; Wu, Chao; Xiang, Tao; Fang, BinIEEE Transactions on Broadcasting, 2023, 69(2): 369-377.被引 8
