An Optimized Rate Control Algorithm for H.264 under Low Delay Constraint

摘要

It is very critical to determine a suitable quantity parameter (QP) in the rate control of video compression. The first frame QP of a GOP (group of picture) affects subsequent frame QP of this GOP, therefore, it has an important impact on the compression video quality of entire video sequences. However, the existing rate control algorithms do not pay much attention to this issue. In this paper, the initial QP prediction (IQPP) algorithm that can predict the initial QP precisely is proposed based on statistics of encoded frames and frame complexity, and the remaining QP prediction (RQPP) algorithm adopts Lagrangian optimization techniques to choose the appropriate QP for I-frame of other GOPs. Our experimental results show the proposed algorithm greatly enhances the performance of rate control and rate-distortion optimization (RDO). Furthermore, the average Peak Signal-to-Noise Ratio (PSNR) is improved and skipped frames are also reduced compared to traditional method under low delay constraint.