Paper Info
Title
Evaluation of full-reference objective video quality metrics on high efficiency video coding.
Abstract
For the purpose of automatic video quality evaluation, Peak Signal-to-Noise Ratio (PSNR) has been the most well-known full reference quality metric since a long time. Improving on PSNR, several other pixel-based quality metrics have been developed, namely Structural SIMilarity (SSIM), Multi Scale-SSIM (MS-SSIM), and Video Quality Metric (VQM). The goal of these objective video quality metrics is to replace time-consuming and expensive subjective quality assessment experiments. These alternative objective metrics have already been evaluated on several video compression schemes, such as MPEG-2 and H. 264/AVC, transported over different kinds of network protocols and under a large variation of network characteristics. In January 2013, the successor of the Advanced Video Coding (H. 264/AVC) standard, named High Efficiency Video Coding (HEVC), has been finalized. Although HEVC is still a block based hybrid video compression standard, some radical changes are made to subjectively improve the compression efficiency compared to H. 264/AVC. Until now, the alternative quality metrics have never been evaluated on this new compression scheme. Therefore, in this paper, we analyze the difference in performance of these full reference metrics. Based on subjective evaluations, a performance analysis is presented which shows the validity of these models when applied to HEVC compressed video content. I. INTRODUCTION
Year
Venue
Keywords
2013
IM
data compression,performance evaluation,protocols,video coding,H.264/AVC standard,HEVC compressed video content,MPEG-2,MS-SSIM,PSNR,VQM,advanced video coding standard,automatic video quality evaluation,block based hybrid video compression standard,full-reference objective video quality metric evaluation,high efficiency video coding,multiscale-SSIM,network characteristics,network protocols,peak signal-to-noise ratio,pixel-based quality metrics,structural similarity
Field
DocType
Citations 
Data mining,Computer science,Multiview Video Coding,PEVQ,Artificial intelligence,Video quality,Rate–distortion optimization,Video compression picture types,Distributed computing,Scalable Video Coding,Context-adaptive binary arithmetic coding,Computer vision,Subjective video quality
Conference
0
PageRank 
References 
Authors
0.34
0
7
Name
Order
Citations
PageRank
Glenn Van Wallendael113723.28
Sebastiaan Van Leuven210313.08
Jan De Cock338849.57
Peter Lambert453867.24
Rik Van de Walle52040238.28
Nicolas Staelens617413.53
Piet Demeester73471363.78