Subjective video quality assessment of VP9 compression standard for full HD resolution

Interest in new multimedia services has rapidly increased over last years. This concern goes hand in hand with demand for higher bandwidth which leads to need to develop new compression standards as VP9 or H.265/HEVC. This results in requirements for video quality assessment as one main part of the...

Full description

Saved in:
Bibliographic Details
Published in2016 International Symposium on Performance Evaluation of Computer and Telecommunication Systems (SPECTS) pp. 1 - 5
Main Authors Uhrina, M., Bienik, J., Vaculik, M., Voznak, M.
Format Conference Proceeding
LanguageEnglish
Published The Society for Modeling and Simulation International 01.07.2016
Subjects
ACR
Bit rate
DSCQS
DSIS
Multimedia communication
Observers
Quality assessment
subjective assessment
video quality
Online AccessGet full text

Cover

More Information
Summary:Interest in new multimedia services has rapidly increased over last years. This concern goes hand in hand with demand for higher bandwidth which leads to need to develop new compression standards as VP9 or H.265/HEVC. This results in requirements for video quality assessment as one main part of the multimedia stream. Compression together with transmission link imperfection are two major factors that influence the quality. This paper deals with assessment of impact of VP9 compression standard on the video quality using different subjective metrics. The evaluation is done for two types of sequences with Full HD resolution depending on content. The main outcome obtained from the results is the determination of the threshold in the means of point where the quality do not rise so swift. This threshold was set between 5 and 7 Mbps. After this threshold the observers did not evaluate the quality of both sequence with quite higher marks which means there is no need to compress the sequences to higher bitrates than to 7 Mbps. These results are part of a new model that is still being created and will be used for predicting the video quality in networks based on IP.
DOI:10.1109/SPECTS.2016.7570522