Deep CNN-Based Blind Image Quality Predictor

By Jongyoo Kim, Anh-Duc Nguyen, Sanghoon Lee. IEEE Transactions on Neural Networks and Learning Systems, 2019

306 citations recorded by the source corpus.

Image recognition based on convolutional neural networks (CNNs) has recently been shown to deliver the state-of-the-art performance in various areas of computer vision and image processing. Nevertheless, applying a deep CNN to no-reference image quality assessment (NR-IQA) remains a challenging task due to critical obstacles, i.e., the lack of a training database. In this paper, we propose a CNN-based NR-IQA framework that can effectively solve this problem. The proposed method—deep image quality assessor (DIQA)—separates the training of NR-IQA into two stages: 1) an objective distortion part and 2) a human visual system-related part. In the first stage, the CNN learns to predict the objective error map, and then the model learns to predict subjective score in the second stage. To complement the inaccuracy of the objective error map prediction on the homogeneous region, we also propose a reliability map. Two simple handcrafted features were additionally employed to further enhance the accuracy. In addition, we propose a way to visualize perceptual error maps to analyze what was learned by the deep CNN model. In the experiments, the DIQA yielded the state-of-the-art accuracy on the various databases.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/4b1f961ae1fac044c23c51274d92d0b26722f877

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

1. Guangming Shi
2. Weisheng Dong
3. Hancheng Zhu
4. Jinjian Wu
5. Jinqiu Sun
6. Yanning Zhang
7. Xin Ge
8. Leida Li

In Media Technology

1. Guangming Shi
2. Weisheng Dong
3. Hancheng Zhu
4. Jinjian Wu
5. Jinqiu Sun
6. Yanning Zhang
7. Xin Ge
8. Leida Li

Citations in the corpus

• MetaIQA: Deep Meta-Learning for No-Reference Image Quality Assessment (2020, methodology): “The five handcrafted feature-based NR-IQA methods are BLIINDSII [39], BRISQUE [32], ILNIQE [56], CORNIA [54] and HOSA [49], while the six deep learning-based NRIQA methods are BIECON [18], MEON [30], WaDIQaMNR [2], DistNet-Q3 [3], DIQA [19] and NSSADNN [52].”
• Blindly Assess Image Quality in the Wild Guided by a Self-Adaptive Hyper Network (2020, methodology): “In [16,23,31], error map of the distorted image is learned to guide quality prediction, the approaches learning error map include training with residual error [16], with quality map calculated from FR-IQA methods [31] and with GAN generated image references [23].”