End-to-End Blind Image Quality Assessment Using Deep Neural Networks

We propose a multi-task end-to-end optimized deep neural network (MEON) for blind image quality assessment (BIQA). MEON consists of two sub-networks-a distortion identification network and a quality prediction network-sharing the early layers. Unlike traditional methods used for training multi-task...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on image processing Jg. 27; H. 3; S. 1202 - 1213
Hauptverfasser: Ma, Kede, Liu, Wentao, Zhang, Kai, Duanmu, Zhengfang, Wang, Zhou, Zuo, Wangmeng
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States IEEE 01.03.2018
Schlagworte:
Blind image quality assessment
deep neural networks
generalized divisive normalization
gMAD competition
Image coding
Image quality
multi-task learning
Neural networks
Nonlinear distortion
Training
ISSN:1057-7149, 1941-0042, 1941-0042
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We propose a multi-task end-to-end optimized deep neural network (MEON) for blind image quality assessment (BIQA). MEON consists of two sub-networks-a distortion identification network and a quality prediction network-sharing the early layers. Unlike traditional methods used for training multi-task networks, our training process is performed in two steps. In the first step, we train a distortion type identification sub-network, for which large-scale training samples are readily available. In the second step, starting from the pre-trained early layers and the outputs of the first sub-network, we train a quality prediction sub-network using a variant of the stochastic gradient descent method. Different from most deep neural networks, we choose biologically inspired generalized divisive normalization (GDN) instead of rectified linear unit as the activation function. We empirically demonstrate that GDN is effective at reducing model parameters/layers while achieving similar quality prediction performance. With modest model complexity, the proposed MEON index achieves state-of-the-art performance on four publicly available benchmarks. Moreover, we demonstrate the strong competitiveness of MEON against state-of-the-art BIQA models using the group maximum differentiation competition methodology.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1057-7149
1941-0042
1941-0042
DOI:10.1109/TIP.2017.2774045