Skip to main navigation Skip to search Skip to main content

COME for No-Reference Video Quality Assessment

  • University of Chinese Academy of Sciences
  • School of Computer Science and Technology, Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Nowadays, the issue of objective Video Quality Assessment (VQA) has been extensively studied. In this paper, we present an effective general-purpose VQA method named COnvolutional neural network and Multi-regression based Evaluation (COME). It requires no referred lossless video and is universal for non-specific types of distortion. A modified 2D convolutional neural network is introduced to learn the spatial features at frame level. At the same time, the motion information is extracted as temporal features at sequence level. And a multi-regression model is proposed to comprehensively assess the final video quality according to human's psychological perception. The proposed method is tested on two commonly used databases with numerous kinds of distortions. The experimental results show that the proposed COME method is comparable with most popular full-reference VQA methods.

Original languageEnglish
Title of host publicationProceedings - IEEE 1st Conference on Multimedia Information Processing and Retrieval, MIPR 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages232-237
Number of pages6
ISBN (Electronic)9781538618578
DOIs
StatePublished - 26 Jun 2018
Externally publishedYes
Event1st IEEE Conference on Multimedia Information Processing and Retrieval, MIPR 2018 - Miami, United States
Duration: 10 Apr 201812 Apr 2018

Publication series

NameProceedings - IEEE 1st Conference on Multimedia Information Processing and Retrieval, MIPR 2018

Conference

Conference1st IEEE Conference on Multimedia Information Processing and Retrieval, MIPR 2018
Country/TerritoryUnited States
CityMiami
Period10/04/1812/04/18

Keywords

  • 2D convolutional neural network
  • AlextNet
  • multi regression model
  • spatial and temporal feature
  • video quality assessment

Fingerprint

Dive into the research topics of 'COME for No-Reference Video Quality Assessment'. Together they form a unique fingerprint.

Cite this