一种基于率失真优化的关键参考帧选择算法

A Novel Key Frame Reference Picture Selection Algorithm Based on Rate-Distortion Optimization

为抑制压缩视频流的传输错误扩散，提出了一种新的基于关键帧的无反馈参考帧选择算法，通过采用一种适用各种掩盖策略的端到端失真度估计方法，评价出选择不同参考帧情况下的解码器失真度期望，在率失真框架内选择最合适的参考帧，使解码端的失真度达到最小．基于H．263＋TMN8测试模型和IP包丢失的模拟实验表明，提出的率失真优化关键参考帧算法比传统的预测编码方法，PSNR值大约提高了1～3dB．当数据丢包率严重时，优势尤为明显．

To restrain the video transmission error propagation over error prone environment, a new non-feedback reference picture selection algorithm based on key frame was proposed. By using a general end-to-end distortion model that is suitable for various concealment methods, the distortion expectation of decoder is estimated approximately for different reference pictures selection at encoder. Based on rate-distortion optimization rule, the best suitable reference pic- ture is chosen to minimize the decoder distortion. Simulation results in H. 263 ＋ test model TMN8 and over IP lost environment show that the proposed algorithm can increase about 1 ～ 3 dB performance gains comparing with traditional prediction algorithm for different QCIF video sequences.