基于x264编码的码率控制模型优化方法

Optimization Method for Rate Control Model Based on x264 Encoding

在视频的直播过程中,传统的编码模式难以在保证视频质量的同时提高信道利用率,导致客户端无法实现流畅地播放。基于x264提出CVBV(ConstantVideo Buffering Verifier)码率控制模型,融合CRF和视频缓冲校验两种模型优势,通过在切片内部使用缓冲区限制码率波动,根据宏区块平滑程度调整量化参数和偏移量,在切片层级实现码率控制模型的优化。通过对不同复杂度视频的不同控制模式及不同编码方式进行对比实验,测试切片视频序列的峰值信噪比(PSNR)、结构相似性(SSIM)和码率波动。实验结果表明,对比x264原有的四种码率控制模型。上述模型在视频质量和率失真指标上基本持平或略有提升(PSNR达到47.91dB,SSIM达到0.993),但视频压缩率有明显提升,提升9.23%。

During the live broadcast of video,it is difficult for the traditional encoding mode to improve the channel utilization rate while ensuring the quality of the video,resulting in the failure of the client to achieve smooth playback.Based on x264,this paper proposes a Constant Video Buffering Verifier(CVBV)rate control model,which combines the advantages of CRF and video buffer verification.By using the buffer in the slice to limit the rate fluctuations,the quantization parameters and offset were adjusted according to the smoothness of the macro block,and the optimization of the rate control model at the slice level was realized.Through comparative experiments on different control modes and different encoding methods of videos of different complexity,the peak signal-to-noise ratio(PSNR),structural similarity(SSIM)and bit rate fluctuations of sliced video sequences were tested.Experimental results show that compared with the original four rate control models of x264,the model in this paper is basically the same or slightly improved in terms of video quality and rate-distortion indicators(PSNR reaches 47.91dB,SSIM reaches 0.993),but the video compression rate is significantly improved,an increase of 9.23%.