一种用于层间预测的自适应插值方法

Adaptive interpolation scheme for inter-layer prediction

为消除可分级视频编码中的层间冗余,提出了一种用于可分级视频编码层间预测的自适应插值方法。用于亮度信息的插值滤波器根据量化参数从3种滤波器中选择,包括双线性滤波器、双三次滤波器和六抽头滤波器;对色度信息进行插值时,使用双线性滤波器代替联合可分级视频编码模型(JSVM)使用的六抽头滤波器。实验结果表明,与JSVM所使用的固定插值滤波器相比,提出的自适应插值方法性能提高0.05dB,同时改善了重建视频中快速运动物体的主观质量。

To eliminate the redundancies among layers in scalable video coding, an adaptive interpolation scheme for inter-layer prediction in scalable video coding was presented. Firstly, the interpolation filter used for luma information was adaptively selected among the three filters which were bilinear, bicubic, and 6-tap filter based on the quantization parameter. Secondly, the bilinear filter was used for chroma informa- tion instead of the 6-tap filter used in JSVM （Joint Scalable Video Model）. Experimental results show that, compared with the fixed interpolation scheme used in JSVM, the performance of the proposed scheme increases by 0.05 dB, and the subjective quality of the fast-moving objects in reconstructed video is improved.