Multilevel Feature Fusion for End-to-End Blind Image Quality Assessment

Xuting LAN, Mingliang ZHOU*, Xueyong XU, Xuekai WEI, Xingran LIAO, Huayan PU, Jun LUO, Tao XIANG, Bin FANG, Zhaowei SHANG

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

11 Citations (Scopus)

Abstract

In this paper, a framework based on two feature extraction networks and a multilevel feature fusion (MFF) network is proposed. Multilevel degradation features can be obtained through this method, and combined with the human visual perception system, the local and global feature information contained in these features can be captured, which is conducive to the prediction of distorted images. First, a restored image approximating a reference image is generated by a restorative generative adversarial network (GAN). Furthermore, the multilevel degradation features of distorted images and the restored image features are extracted by EfficientNet. Second, the features extracted by EfficientNet are input into the MFF network and are fully expressed by the top-down, bottom-up and third edge joining methods. Moreover, the features provide more low-level details and high-level semantic features for the prediction of image quality scores. In addition, after the MFF stage, the framework calculates the score of each branch feature and obtains the average quality score. Experimental results show that our method achieves greatly improved prediction accuracy and performance on five standard databases.

Original languageEnglish
Pages (from-to)801-811
Number of pages11
JournalIEEE Transactions on Broadcasting
Volume69
Issue number3
Early online date4 Apr 2023
DOIs
Publication statusPublished - Sept 2023
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 1963-12012 IEEE.

Keywords

  • blind image quality assessment
  • Feature extraction
  • multilevel feature fusion

Fingerprint

Dive into the research topics of 'Multilevel Feature Fusion for End-to-End Blind Image Quality Assessment'. Together they form a unique fingerprint.

Cite this