基于率失真优化的可伸缩视频编码码率分配

RATE-DISTORTION OPTIMIZATION BASED RATE ALLOCATION IN SCALABLE VIDEO CODING

可伸缩视频编码的主要目标是生成空域、时域、质量可伸缩的视频流,其中如何进行有效的码率分配,以截取特定条件下最优的一段码流,对视频信号的网络传输十分重要。在研究现有JVTSVC中码率分配算法的基础上,提出了通过划分FGSNAL单元建立其近似的率失真特性曲线,从中选择构成凸壳的点作为率失真最优的候选截断点,并以此进行优化的码率分配的方法。在参考软件JSVM上的实现表明,对标准测试集中的序列截取相同需求条件下的码流时,相对于JVTSVC中现有率失真优化的码率分配方法,新方法的解码视频质量均有进一步的提高。

The main target of scalable video coding is to generate bit-stream with spatial, temporal and SNR scalability. When bit-stream is transmitted on the network, it is very important to allocate bit-rate effectively in order to get an optimal bit-stream under a certain condition. A new rate allocation method is proposed based on the study on the existing rate allocation algorithm in JVT SVC. Firstly, FGS NAL unit is divided into several fragments, and an approximate rate-distortion （R-D） curve of each FGS NAL unit is established. Then, the R-D points lying on the convex hull of curve are chosen as R-D optimization truncation points, according to which rate allocation is optimized. The experiment is carried on the test sequences of the standard set based on the reference software of JSVM. When extracting the bit-stream with the same requirement, compared with the existing method of R-D optimized rate allocation in JVT SVC, the proposed method achieves more effective rate allocation, and the quality of decoded sequence is further improved.