Sub-Sampling based Double Description Coding with Pixel-wise error concealment for efficient H.264 Video Transmission

碩士 === 國立雲林科技大學 === 電機工程系碩士班 === 93 === How to get a balance among the bit-rate, the power-consumption, and the error robustness is a very difficult problem in the wireless video transmission field. After analyzing many H.264 error resilience tools, we find a common defect which the bit-rate will dr...

Full description

Bibliographic Details
Main Authors: Jung-chi Wang, 王榮諆
Other Authors: Day-fann Shen
Format: Others
Language:zh-TW
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/46354005796398827132
Description
Summary:碩士 === 國立雲林科技大學 === 電機工程系碩士班 === 93 === How to get a balance among the bit-rate, the power-consumption, and the error robustness is a very difficult problem in the wireless video transmission field. After analyzing many H.264 error resilience tools, we find a common defect which the bit-rate will drastically increase with the robustness of error resistance and proposed an Average 3-point sub-sample(A3_45) based multiple description coding(MDC) technique to improve the compression capability by saving the inter-pixel redundancy rather than reduce it in the ordinary spatial sub-sample method. Besides that, the characteristic of the Average 3-point sub-sample(A3_45) based multiple description coding(MDC) technique is suitable for Pixel-wise error concealment which can have more exquisite video than Block-wise error concealment. In addition, we find that even with only two compression descriptions, we can efficiently decrease the power-consumption without sacrificing the visual quality. As experiment results show, we can verify that H.264 SS MDC proposed in this thesis have almost 50% lower bit-rate than direct sub-sample method and efficiently decrease up to half the power-consumption. Especially when packets are lost, pixel-wise error concealment can cause much better visual quality.