[Zurück]

J. Colom Ikuno, L. Superiori, M. Rupp:
"Quality Estimation of Damaged H.264/AVC sequences";
Vortrag: Redzur 2009, 3rd International Workshop on Speech and Signal processing, Bratislava; 24.09.2009; in: "Redzur 2009, 3rd International Workshop on Speech and Signal processing", Nakladateľstvo STU Bratislava in cooperation with Zdruţenie pouţívateľov telekomunikácií Slovenska, Bratislava (2009), ISBN: 978-80-227-3137-9; S. 8 - 11.

In this paper we propose a subjective quality predictor for
damaged low resolution H.264/AVC encoded sequences. The
predictor utilizes the information relayed by a visual artifacts
detection algorithm working at pixel level and is supported by
a syntax analysis error detector working at bit level. The algorithm
is capable of predicting the Mean Opinion Score (MOS)
of the received damaged sequence. Preliminary results show
a strong correlation between the predicted data and subjective
measurement tests with volunteers. Such a method could be
used over video transmissions in error-prone channels, such
as in wireless systems to adjust the transmission to the estimated
received subjective quality.

http://publik.tuwien.ac.at/files/PubDat_177611.pdf

Projektleitung Markus Rupp:
Mobilkom