Modeling DCT coefficients for fast video encoding

Abstract

Digital video coding standards such as H.263 and MPEG are becoming more and more important for multimedia applications. Due to the huge amount of computations required, there are significant efforts to speed up the processing of video encoders. Previously, the efforts were mainly focused on the fast motion-estimation algorithm. However, as the motion-estimation algorithm becomes optimized, to speed up the video encoders further we also need to optimize other functions such as the discrete cosine transform (DCT) and inverse DCT (IDCT). In this paper, we propose a theoretical model for DCT coefficients. Based on the model, we develop an adaptive algorithm to reduce the computations of DCT, IDCT, quantization, and inverse quantization. We also present a fast DCT algorithm to speed up the calculations of DCT further when the quantization step size is large. We show, by simulations, that significant improvement in the processing speed can be achieved with negligible video-quality degradation, We also implement the algorithm in a real-time PC-based platform to show that it is effective and practical.