Control of video processing algorithms based on measured perceptual quality characteristics

Kalpana Seshadrinathan, Jorge E. Caviedes

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

1 Scopus citations

Abstract

Most consumer electronic devices include a suite of video post-processing algorithms that process the output of the video decoder to drive the video display. In this paper, we propose the use of non-reference video quality metrics to improve the Quality of Experience (QoE) of the user on-the-fly by controlling the parameters of the video post-processing pipeline. We utilize non-reference and real-time measurements of perceptual contrast, sharpness and noise to control denoising and contrast enhancement algorithms in the video post-processing pipeline. We propose methods to ensure consistency of the post-processing across the different frames of the video to avoid creation of flickering artifacts due to variable post-processing. We show that video quality can be improved by identifying the specific quality characteristics that are altered by a video post-processing method and using measurements of these characteristics to improve overall user QoE.

Original languageEnglish (US)
Title of host publication2012 IEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2012, Proceedings
Pages177-180
Number of pages4
DOIs
StatePublished - Jul 2 2012
Externally publishedYes
Event2012 IEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2012 - Santa Fe, NM, United States
Duration: Apr 22 2012Apr 24 2012

Publication series

NameProceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation

Other

Other2012 IEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2012
Country/TerritoryUnited States
CitySanta Fe, NM
Period4/22/124/24/12

Keywords

  • control
  • no reference
  • post processing
  • quality metrics
  • video quality

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Control of video processing algorithms based on measured perceptual quality characteristics'. Together they form a unique fingerprint.

Cite this