OBJECTIVE ASSESSMENT OF MPEG-VIDEO QUALITY: A NEURAL-NETWORK APPROACH

Paolo Gastaldo(1), Stefano Rovetta(2), and Rodolfo Zunino(1)

(1) DIBE - Dept. Biophysical and Electronic Engineering
University of Genoa
Via all'Opera Pia 11a - 16145 Genova - Italy
e-mail: {gastaldo, zunino}@dibe.unige.it

(2) INFM, DISI - Dept. of Computer and Information Sciences
University of Genoa
Via Dodecaneso 35, 16146 Genova – Italy
e-mail: ste@disi.unige.it


ABSTRACT

The increasing use of compression standards in broadcasting digital
TV has raised the need for established criteria to measure perceived
quality. This paper presents a methodology using Circular
Back-Propagation (CBP) neural networks for the objective quality
assessment of MPEG video streams. Objective features are continuously
extracted from compressed video streams on a frame-byframe basis; they
feed the CBP network estimating the corresponding perceived quality.
The resulting adaptive modeling of subjective perception supports
a real-time system for monitoring displayed video quality.