Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wild
By Weixia Zhang, Kede Ma, Guangtao Zhai, Xiaokang Yang. IEEE Transactions on Image Processing, 2020
368 citations recorded by the source corpus.
Performance of blind image quality assessment (BIQA) models has been significantly boosted by end-to-end optimization of feature engineering and quality regression. Nevertheless, due to the distributional shift between images simulated in the laboratory and captured in the wild, models trained on databases with synthetic distortions remain particularly weak at handling realistic distortions (and vice versa). To confront the cross-distortion-scenario challenge, we develop a unified BIQA model and an approach of training it for both synthetic and realistic distortions. We first sample pairs of images from individual IQA databases, and compute a probability that the first image of each pair is of higher quality. We then employ the fidelity loss to optimize a deep neural network for BIQA over a large number of such image pairs. We also explicitly enforce a hinge constraint to regularize uncertainty estimation during optimization. Extensive experiments on six IQA databases show the promise of the learned method in blindly assessing image quality in the laboratory and wild. In addition, we demonstrate the universality of the proposed training strategy by using it to improve existing BIQA models.
Subfields: Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/750ca6e5b634bba1be0c049b568dfb6dac06e133
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Xiaokang Yang
- Guangtao Zhai
- Kede Ma
- Weixia Zhang
- Kris M. Kitani
- S. Alireza Golestaneh
- Saba Dadsetan
- Ying Wei
Citations in the corpus
- Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspective (2023, methodology): “Similar to [78], we choose to jointly train a BIQA model on M (M ≥ 2) IQA datasets.”
- No-Reference Image Quality Assessment via Transformers, Relative Ranking, and Self-Consistency (2021, methodology): “[63] used continuous ranking information from MOSs and variances between the subjective scores.”
- Exploring Video Quality Assessment on User Generated Contents from Aesthetic and Technical Perspectives (2022, neutral): “5647 UNIQUE (2021) [76] fused ( [15, 73–75]) 0.”