首页|ERP全景视频VVC帧内编码CU快速划分算法

ERP全景视频VVC帧内编码CU快速划分算法

扫码查看
针对等矩柱状投影(equirectangular projection,ERP)全景视频多功能视频编码(versatile video coding,VVC)帧内编码复杂度过高的问题,提出一种编码单元(coding unit,CU)快速划分算法.根据ERP采样特性,将编码帧分为不同纬度区域.基于不同纬度区域CU四叉树深度的分布特性和相邻CU的相关性,对当前CU的划分模式进行提前终止决策;利用梯度差异评估当前CU纹理特性,跳过冗余的水平或垂直划分模式.针对纹理模糊CU,通过纬度采样权重加权的二次比较,判断是否跳过垂直划分模式;利用二维哈尔小波变换系数评估当前CU子块间的差异,判断是否跳过三叉树划分模式.实验结果表明,在全帧内模式下,与VVC官方测试平台相比,所提算法能节省43.85%的编码时间,码率仅增加0.85%,视频质量没有明显下降.
VVC intra-coding CU fast partition algorithm for ERP panoramic video
A coding unit fast partition algorithm is proposed to address the problem of high complexity of intra-frame coding in equirectangular projection versatile video coding panoramic video.Firstly,this paper divided the coding frame into differ-ent latitude regions according to ERP sampling characteristics.Then,this method made an early termination decision for the current CU division mode based on the distribution characteristics of CU quadtree depths in different latitude areas and the correlation of adjacent CUs.This algorithm used the gradient difference to evaluate the current CU texture characteristics to skip the redundant horizontal or vertical division mode and compared the texture fuzzy CUs by weighting the latitude sam-pling weights to judge whether to skip the vertical division mode or not.Finally,this algorithm used the two-dimensional Hal wavelet transform coefficients to evaluate the difference between the current CU sub-blocks to judge whether to skip the ternary tree division mode.Experimental results show that under the full intra-mode,the proposed algorithm reduces enco-ding time by 43.85%compared to the official VVC test platform,with only a 0.85%increase in bitrate and no significant decline in video quality.

ERP panoramic videoCU partitiontexture featuressub-block differences

李强、董阳、赵宇

展开 >

重庆邮电大学通信与信息工程学院,重庆 400065

ERP全景视频 CU划分 纹理特征 子块差异

国家自然科学基金项目重庆市自然科学基金项目

62176035cstc2021jcyjbshX0206

2024

重庆邮电大学学报(自然科学版)
重庆邮电大学

重庆邮电大学学报(自然科学版)

CSTPCD北大核心
影响因子:0.66
ISSN:1673-825X
年,卷(期):2024.36(5)