An improved side information generation for distributed video coding

doi:10.1016/j.aeue.2013.08.005

An improved side information generation for distributed video coding

2014-03-01

6

PRE

AI

DOI：10.1016/j.aeue.2013.08.005

Original

Original request for help

Save

Abstract

En 中文

This paper presents a side information (SI) scheme for distributed video coding based on multilayer perceptron. The suggested scheme predicts a Wyner-Ziv (WZ) frame from two decoded key frames. The network is trained offline using patterns from different standard video sequences with varied motion characteristics to achieve generalization. The proposed scheme is simulated along with other standard video coding schemes. Performance comparisons have been made with respect to training convergence, rate distortion (RD), peak signal to noise ratio (PSNR), number of requests per SI frame, decoding time requirement, etc. In general, it is observed that the proposed scheme has a superior SI frame generation capability as compared to its competent schemes. (C) 2013 Elsevier GmbH. All rights reserved.

Keywords:

Distributed video coding

Side information

Multi layer perceptron

Rate distortion

Decoding time requirement

An improved side information generation for distributed video coding

Abstract

Journal

Organization