基于视频时空相关性的帧内预测模式抉择

提出了一种帧内预测模式抉择方法,利用视频图像的时间相关性和空间相关性,选择部分预测模式对图像块进行帧内预测处理和编码,减少编码的计算复杂度,保持其原有的编码效率。同时提出了一种复杂度分级方法,通过准确控制帧内预测的复杂度,以匹配通信终端有限的计算资源。该方法在AVS-M平台中进行了验证。

Intra-prediction mode decision based on video temporal and spatial correlation

An intra-prediction mode-decision method was proposed. Based on the inherent spatial correlation within a single frame and temporal correlation between successive frames, part of intra-prediction modes were chosen to determine the best prediction mode for image block, which resulted in directly reducing the computational complexity and maintaining high coding efficiency. In addition, a complexity-scalable technique was presented to control the complexity of intra prediction, by which the video coder was matched to the limited computational resource of communication equipments. The proposed methods are verified in AVS-M software platform.