Network-Adaptive Video Communication Using Packet Path Diversity and Rate-Distortion Optimized Reference Picture Selection

Authors: Liang, Yi¹; Setton, Eric²; Girod, Bernd³

Source: The Journal of VLSI Signal Processing, Volume 41, Number 3, November 2005 , pp. 345-354(10)

Publisher: Springer

Buy & download fulltext article:

Price: $47.00 plus tax (Refund Policy)

Abstract:

In this paper, we present error-resilient Internet video transmission using path diversity and rate-distortion optimized reference picture selection. Under this scheme, the optimal packet dependency is determined adapting to network characteristics and video content, to achieve a better trade-off between coding efficiency and forming independent streams to increase error-resilience. The optimization is achieved within a rate-distortion framework, so that the expected end-to-end distortion is minimized under the given rate constraint. The expected distortion is calculated based on an accurate binary tree modeling with the effects of channel loss and error concealment taken into account. With the aid of active probing, packets are sent across multiple available paths according to a transmission policy which takes advantage of path diversity and seeks to minimize the loss rate. Experiments demonstrate that the proposed scheme provides significant diversity gain, as well as gains over video redundancy coding and the NACK mode of conventional reference picture selection.

Keywords: Path diversity; network-adaptive video coding; error-resilient video coding; source-channel coding; reference picture selection; low-latency; rate-distortion optimization; video streaming

Document Type: Research article

DOI: http://dx.doi.org/10.1007/s11265-005-4157-x

Affiliations: 1: Email: yiliang@stanfordalumni.org 2: Email: esetton@stanford.edu 3: Email: bgirod@stanford.edu

Publication date: 2005-11-01