Error resilient video streaming algorithms based on Wyner-Ziv coding
Abstract
Compressed video is very sensitive to channel errors. A few bit losses can derail the entire decoding process. Therefore, protecting compressed video is always necessary for reliable visual communications. In recent years, Wyner-Ziv lossy coding has been applied to error resilience and achieved superior improvement over conventional techniques. In our work, we first proposed an unequal error protection method for protecting data elements in a video stream, via a Wyner-Ziv encoder that consists of a coarse quantizer and a Turbo coder based lossless Slepian-Wolf encoder. Data elements that significantly impact the visual quality of decoded video, such as modes and motion vectors as used by H.264, are protected more by assigning a higher parity data rate than the coarsely quantized transform coefficients. This results in an improvement in the quality of the decoded video when the transmitted data was corrupted by the transmission errors, than the results obtained by using equal error protection. In our later work, we described an improved technique by adapting the parity bit rates of the protected video information to the video content. The parity data rates are assigned to the motion information and the transform coefficients according to their impact on the visual quality of each frame. This results in an efficient way of improving the quality of the decoded video when it has been corrupted by transmission errors. We also proposed a feedback aided error resilience technique, based on Wyner-Ziv coding. By utilizing feedback regarding current channel packet-loss rates, turbo coders can adaptively adjust the amount of parity bits needed for correcting corrupted slices at the decoder. This results in an efficient usage of the bit budget for Wyner-Ziv coding while maintaining good quality decoded video when the data has been corrupted by transmission errors. Finally we studied the case of combining the content adaptive function at the encoder side and utilizing the feedback of the current channel packet losses conveyed from the decoder side to adjust the protection level of the protection video information at the same time. A new parity data rates assignment table is given. The experiment results showed the improvement on the rate distortion performance of this combined method, comparing to the case of utilizing the content adaptive function or the decoder side feedback information individually. An analysis and modeling of the system have also been studied, which aim to find the theoretical basis for finding the optimal data rate allocation between the primary coding and the Wyner-Ziv coding.
Degree
Ph.D.
Advisors
Delp, Purdue University.
Subject Area
Electrical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.