Robustness of compressed high definition video to transmission packet loss

When comparing video compression algorithms, the most common performance indicator is the rate-distortion dependency, i.e. how effective is the algorithm in reducing the source video bit rate while meeting a certain quality measure, such as a given peak signal-to-noise ratio (PSNR). This paper compares two of the most widespread algorithms in a different manner: the compressed bit stream is subject to simulated erroneous transmission, and the impact of missing data on the decoded video at the receiver end is measured. To objectively and repeatably measure the quality of the decoded video subject to errors, PSNR is used.