Packet loss protection method based on picture level adaptive frame /field coding (PAFF)was presented. Firstly,the end-to-end rate-distortion analysis for PAFF on the current frame was performed. Secondly,in order to evaluate the severity of error propagation in the following frames,the error propagation intensity and human visual quality sensitivity of different areas were taken into consideration. It was followed by the quantification of relative importance. Finally,the proper coding mode was chosen utilizing an unequal comparison procedure. The simulation results show that the proposed method can improve peak signal-to-noise ratio (PSNR) up to 0. 9 dB and 1. 6 dB comparing with the field only and the dispersed flexible macro-block ordering (FMO)only methods respectively.
为了提高视频质量评价的精确度,降低计算复杂度,提出了一种双域无参考视频质量评价算法.在压缩域,利用从码流中提取的编码信息,结合视频内容复杂度来计算视频相似度;在像素域,结合人类视觉系统的特性,通过块效应和模糊效应评估来计算视频失真度;最后,综合视频相似度和失真度得出视频客观质量.文中选用LIVE Video QualityDatabase对算法效果进行了评估,结果表明:该算法的预测精确性和单调性分别达到0.801 3和0.786 1,优于全部10种对比的视频客观质量评价方法;同时,该算法具有较低的计算复杂度,可以满足实时应用要求.
