An efficient error resilience scheme based on Wyner-Ziv coding for region-of-interest protection of wavelet based video transmission.
Zhuo, Xue and Loo, Jonathan and Cosmas, John and Yip, Pik Yee (2008) An efficient error resilience scheme based on Wyner-Ziv coding for region-of-interest protection of wavelet based video transmission. WSEAS Transaction on Circuits and Systems, 7 (4). pp. 277-286. ISSN 1109-2734
- Published Version
In this paper, we propose a bandwidth efficient error resilience scheme for wavelet based video transmission over wireless channel by introducing an additional Wyner-Ziv (WZ) stream to protect region of interest (ROI) in a frame. In the proposed architecture, the main video stream is compressed by a generic wavelet domain coding structure and passed through the error prone channel without any protection. Meanwhile, the predefined ROI area related wavelet coefficients obtained after an integer wavelet transform will be specially protected by WZ codec in an additional channel during transmission. At the decoder side, the error-prone ROI related wavelet coefficients will be used as side information to help decoding the WZ stream. Different size of WZ bit streams can be applied in order to meet different bandwidth condition and different requirement of end users. The simulation results clearly revealed that the proposed scheme has distinct advantages in saving bandwidth comparing with fully applied FEC algorithm to whole video stream and in the meantime offer the robust transmission over error prone channel for certain video applications.
|Research Areas:||School of Science and Technology > Computer and Communications Engineering|
School of Science and Technology > Science & Technology
|Permissions granted by publisher:||We acknowledge full permission from WSEAS to include published version. (5/11)|
|Deposited On:||28 Apr 2011 15:47|
|Last Modified:||21 Jul 2014 04:10|
Repository staff 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