Neural network-based full-reference image quality assessment

Sebastian Bosse, Dominique Maniry, Klaus Muller, Thomas Wiegand, Wojciech Samek

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

This paper presents a full-reference (FR) image quality assessment (IQA) method based on a deep convolutional neural network (CNN). The CNN extracts features from distorted and reference image patches and estimates the perceived quality of the distorted ones by combining and regressing the feature vectors using two fully connected layers. The CNN consists of 12 convolution and max-pooling layers; activation is done by a rectifier activation function (ReLU). The overall IQA score is computed by aggregating the patch quality estimates. Three different feature combination methods and two aggregation approaches are proposed and evaluated in this paper. Experiments are performed on the LIVE and TID2013 databases. On both databases linear Pearson correlations superior to state-of-the-art IQA methods are achieved.

Original languageEnglish
Title of host publication2016 Picture Coding Symposium, PCS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509059669
DOIs
Publication statusPublished - 2017 Apr 19
Event2016 Picture Coding Symposium, PCS 2016 - Nuremberg, Germany
Duration: 2016 Dec 42016 Dec 7

Other

Other2016 Picture Coding Symposium, PCS 2016
CountryGermany
CityNuremberg
Period16/12/416/12/7

ASJC Scopus subject areas

  • Media Technology
  • Signal Processing

Fingerprint Dive into the research topics of 'Neural network-based full-reference image quality assessment'. Together they form a unique fingerprint.

  • Cite this

    Bosse, S., Maniry, D., Muller, K., Wiegand, T., & Samek, W. (2017). Neural network-based full-reference image quality assessment. In 2016 Picture Coding Symposium, PCS 2016 [7906376] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PCS.2016.7906376