Share Email Print

Proceedings Paper

Interframe wavelet video coding with operating point adaptation
Author(s): Thomas Rusert
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Contrary to predictive schemes such as hybrid video coding systems, orthonormal transform coding systems are immune to error accumulation in case of desynchronization between encoder and decoder. Therefore, these systems allow for drift-free data adaptation at bit stream level, thus, scalability. In t+2D interframe wavelet video coding, wavelet-based motion-compensated temporal filtering is employed, followed by spatial wavelet decomposition and bit plane coding. This allows for temporal, spatial, and SNR scalability. While motion compensation seems to be essential in this scheme to achieve excellent coding performance, it causes local violation of the orthonormality of the temporal transform. Particularly, motion compensated interframe wavelet systems employ predictive coding for certain occlusion areas. In case of reference mismatch between encoder and decoder, error accumulation occurs in these regions. In this paper we present an approach to adapt the encoder operating point for predictively coded regions, effectively eliminating the reference mismatch adaptively. An iterative algorithm for computation of the decoder reference at the encoder side is presented for t+2D systems. We show that this approach significantly increases overall coding performance, gaining up to 1 dB in PSNR. Furthermore, the optimized quantization algorithm presented in an earlier work can be applied more effectively, leading to more even noise distribution.

Paper Details

Date Published: 18 January 2004
PDF: 12 pages
Proc. SPIE 5308, Visual Communications and Image Processing 2004, (18 January 2004); doi: 10.1117/12.525695
Show Author Affiliations
Thomas Rusert, RWTH-Aachen (Germany)

Published in SPIE Proceedings Vol. 5308:
Visual Communications and Image Processing 2004
Sethuraman Panchanathan; Bhaskaran Vasudev, Editor(s)

© SPIE. Terms of Use
Back to Top