Interdependency Measure and Significance Analysis for H.264 Encoding Parameters

Abstract

The H.264 encoder has input parameters that determine the bit rate and distortion of the compressed video and the encoding complexity. These encoding parameters may significantly affect not only the encoding performance, but also affect the cross-layer design optimization for the video streaming transmission over wireless networks. However, it is very difficult to derive the impact factors of these parameters and of the correlation between these parameters. In this paper, we propose a new approach for dynamic interdependency measure among encoding parameters by using non-additive measure theory. In particular, the Choquet integral model is applied to distinguish the interaction among encoding parameters towards the encoding performance through a set of non-additive measures. Then the most significant effect subset of encoding parameters on the performance metrics of interest for objective function is identified. The simulations show that, by only adjusting the parameters in the optimized subset according to the interdependency measure, the encoding performance can be adapt to variety of video content and channel status. Furthermore, the proposed method can be easily implemented and incorporated in existing cross-layer design optimization.