Abstract:

In order to reduce the coding complexity of surveillance and conference video in High Efficiency Video Coding（HEVC） and reduce the coding time, a fast inter frame Coding Unit（CU） partitioning and Prediction Unit（PU） mode selection algorithm is proposed. The pixel of original video sequence is statistically analyzed to generate virtual background frame which is used to classify the CU. The current CU partitioning and the PU mode type are quickly determined by analyzing the proportions of further-partitioned different classes of CU and their texture complexity. The algorithm for virtual background frame is used to make the decision to reduce the traversal process about CU depth and PU modes in inter coding, thereby achieving the effect of fast inter coding. The experimental results show that, compared with HM16.9, the algorithm can reduce the coding time by 33.31% on average when the Peak Signal-to-Noise Ratio（PSNR） decreased by 0.07 dB and the code rate increased by 1.93%.

Key words: high efficiency video coding, inter coding, virtual background frame, CU partitioning, PU mode selection

摘要：

为降低高性能视频编码（High Efficiency Video Coding，HEVC）中监控和会议视频的编码复杂度，减少编码时间，提出了一种基于虚拟背景帧的监控和会议视频帧间编码的快速编码单元（Coding Unit，CU）划分和预测单元（Prediction Unit，PU）模式选择算法。对原始视频序列进行像素统计分析生成虚拟背景帧，利用生成的虚拟背景帧对CU分类。通过分析不同类别的CU继续划分的比例和其纹理复杂度，快速决策当前CU的划分和PU的模式类型。这种利用虚拟背景帧做判决的算法通过减少CU深度和PU模式遍历而降低编码复杂度，达到快速帧间编码的效果。实验结果表明，与HM16.9相比，该算法在峰值信噪比（Peak Signal-to-Noise Ratio，PSNR）平均下降0.07 dB和码率平均增加1.93%的情况下，能平均减少33.31%的编码时间。

关键词: 高性能视频编码, 帧间编码, 虚拟背景帧, CU划分, PU模式选择