Efficient rate-quantization model for frame level rate control in spatially scalable video coding

This paper addresses the quantization parameter (QP) selection problem in H.264 spatially scalable video coding (SVC). For frame level rate control in SVC, it is important to have an accurate QP selection scheme such that the target bit rate of each coding layer will be achieved. In this paper, we present an adaptive rate-quantization (R-Q) model to select the appropriate QP for each inter frame in spatial enhancement layers according to the target bit rate. The proposed algorithm introduces an efficient coding complexity estimation method by taking into consideration the inter-layer dependency between different spatial layers. Based on the coding complexity information, the R-Q model parameters can be adaptively updated. Experimental results demonstrate that compared to the traditional method, the proposed method provides better estimation accuracy for bit rate in terms of target bits mismatch error and thus it is very desirable for H.264/SVC rate control applications.