The Taub Faculty of Computer Science Events and Talks

Visual communication often requires adaptation of the transmission bit-rate to the available channel bandwidth or display characteristics of the end users. In this work, requantization for transrating of MPEG video sequences and JPEG still images is analyzed. We show that both the rate and the distortion of requantized images and video depend mainly on the ratio between the new and the original quantization steps. Our analysis is based on the structure of the quantizer and the Laplace-like distribution of the DCT coefficients in sub-band coding. We also show that the rounding policy has a significant effect on the requantization performance. In addition, we examine a transrating approach that incorporates image downscaling along with requantization. This method is based on the well known coding trade-off between the number of quantization levels and the number of pixels used to code an image. We propose a method that efficiently selects one of the coding modes: requantization or downscaling. We demonstrate how to incorporate the proposed requantization approach into a rate-control system using the Rho model as an example. We show that the resulting bit-rate of the proposed requantization method can be accurately predicted using this model. Our conclusion is that the new approach could be instrumental in achieving a required bit-rate when recompressing still images and video at low distortion, while allowing real-time implementation due to its low computational complexity.

M.Sc Thesis under the supervision of Dr. Moshe Porat.