Blind Image Quality Assessment Using a Deep Bilinear Convolutional Neural Network

By Weixia Zhang, Kede Ma, Jia Yan, Dexiang Deng, Zhou Wang. IEEE transactions on circuits and systems for video technology (Print), 2019

937 citations recorded by the source corpus.

We propose a deep bilinear model for blind image quality assessment that works for both synthetically and authentically distorted images. Our model constitutes two streams of deep convolutional neural networks (CNNs), specializing in two distortion scenarios separately. For synthetic distortions, we first pre-train a CNN to classify the distortion type and the level of an input image, whose ground truth label is readily available at a large scale. For authentic distortions, we make use of a pre-train CNN (VGG-16) for the image classification task. The two feature sets are bilinearly pooled into one representation for a final quality prediction. We fine-tune the whole network on the target databases using a variant of stochastic gradient descent. The extensive experimental results show that the proposed model achieves state-of-the-art performance on both synthetic and authentic IQA databases. Furthermore, we verify the generalizability of our method on the large-scale Waterloo Exploration Database, and demonstrate its competitiveness using the group maximum differentiation competition methodology.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/974ff86cdf8550cffde4d8048609baaac876b563

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

Citations in the corpus

No-Reference Image Quality Assessment via Transformers, Relative Ranking, and Self-Consistency (2021, background): “Zhang et al. [39] leveraged discrete ranking information from images of the same content and distortion but at different levels (degree of distortion) for quality prediction.”
Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspective (2023, methodology): “Being separately trained on each individual dataset with a separate group of parameters, DBCNN [76], HyperIQA [55], and TreS [13] achieve promising performance on all six datasets.”
Exploring CLIP for Assessing the Look and Feel of Images (2022, methodology): “Unlike existing NR-IQA methods ( e.g. , MUSIQ) that train different models for different domains ( e.g. , technical quality, aesthet-3 ics quality), CLIP-IQA + needs to store only two prompts for each domain, signiﬁcantly reducing the storage cost.”
Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wild (2020, methodology): “However, these models remain notoriously weak at crossdistortion-scenario generalization [3].”
MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessment (2022, methodology): “With the success of deep learning and Convolution Neural Network (CNN) in many computer vision tasks, CNN-based methods have significantly outperformed previous approaches in terms of real-world distortions by directly extracting discriminative features from LQ images [1,3,14,20,26,37,39,45,59,61].”
TOPIQ: A Top-Down Approach From Semantics to Distortions for Image Quality Assessment (2023, methodology): “The same as [6], [35], we fitted a 4-parameter”
MetaIQA: Deep Meta-Learning for No-Reference Image Quality Assessment (2020, methodology): “Recent years have witnessed the great success of Deep Convolutional Neural Networks (DCNNs) [14] in many computer vision tasks [4, 5], which has also spawned a number of DCNNs-based NR-IQA approaches [16, 2, 28, 30, 52, 58, 57].”
Internet of Underwater Things and Big Marine Data Analytics—A Comprehensive Survey (2020, background): “Furthermore, the benefits of DNNs in underwater image quality assessment can be assessed in future studies with reference to their overwater counterparts [252].”
FAST-VQA: Efficient End-to-end Video Quality Assessment with Fragment Sampling (2022, background): “Naive approaches such as resizing [17,43] and cropping [14,15] cannot preserve video quality well.”
MUSIQ: Multi-scale Image Quality Transformer (2021, background): “CNN-based IQA models usually resize the input image to a square resolution without preserving the original aspect ratio.”
Q-Align: Teaching LMMs for Visual Scoring via Discrete Text-Defined Levels (2023, methodology): “To address this need, from handcraft approaches (Mit-tal et al., 2013; 2012) to deep-neural-network-based meth-ods (Talebi & Milanfar, 2018; Zhang et al., 2020; Ke et al., 2021), the endeavor to improve the accuracies of visual assessors never stops.”
Blindly Assess Image Quality in the Wild Guided by a Self-Adaptive Hyper Network (2020, methodology): “In [46], a two stream network architecture is introduced to both predict synthetic and authentic image distortions.”
Exploring Video Quality Assessment on User Generated Contents from Aesthetic and Technical Perspectives (2022, background): “Many studies [13, 59, 68] suggest that technical quality perception should consider global semantics to better assess distortion levels.”
Perceptual Quality Assessment of Smartphone Photography (2020, methodology): “These cover a wide range of design philoso-phies, including NSS-based [6, 22, 24, 40], codebook-based [38, 39], and DNN-based [41] models.”