Error-resilient performance of Dirac Video Codec over Packet-Erasure Channel

Tun, Myo and Loo, Jonathan and Cosmas, John (2007) Error-resilient performance of Dirac Video Codec over Packet-Erasure Channel. IEEE Transactions on Broadcasting, 53 (3). pp. 649-659. ISSN 0018-9316

Full text is not in this repository.

Official URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arn...

This item is available in the Library Catalogue

Abstract

Video transmission over the wireless or wired network requires error-resilient mechanism since compressed video bitstreams are sensitive to transmission errors because of the use of predictive coding and variable length coding. This paper investigates the performance of a simple and low complexity error-resilient coding scheme which combines source and channel coding to protect compressed bitstream of wavelet-based Dirac video codec in the packet-erasure channel. By partitioning the wavelet transform coefficients of the motion-compensated residual frame into groups and independently processing each group using arithmetic and Forward Error Correction (FEC) coding, Dirac could achieves the robustness to transmission errors by giving the video quality which is gracefully decreasing over a range of packet loss rates up to 30% when compared with conventional FEC only methods. Simulation results also show that the proposed scheme using multiple partitions can achieve up to 10 dB PSNR gain over its existing un-partitioned format. This paper also investigates the error-resilient performance of the proposed scheme in comparison with H.264 over packet-erasure channel.

Item Type:Article
Research Areas:School of Science and Technology > Computer and Communications Engineering
School of Science and Technology > Science & Technology
Citations on ISI Web of Science:7
ID Code:7828
Useful Links:
Deposited On:03 May 2011 15:07
Last Modified:13 May 2014 15:37

Repository staff and depositor only: item control page

Full text downloads (NB count will be zero if no full text documents are attached to the record)

Downloads per month over the past year