Abstract: Depth image based rendering is one of the key technologies in three-dimensional system. However, coding distortion in depth map will affect the following performance of virtual view synthesis. In this paper, a depth map post-processing method is proposed for virtual view synthesis. In the method, the filtering window size for each pixel is established based on the intersection of confidence intervals rule. The depth value for each pixel within the filtering window is estimated by using least square polynomial approximation. A weighted mode filtering is applied to the estimated depth map, and the filtered depth map is obtained. Experimental results show that, the proposed method can accurately recover the depth information from the decoded depth map, and promote the performance of virtual view synthesis.

Key words: depth map post-processing, intersection of confidence intervals, least square polynomial approximation, weighted mode filtering

摘要： 基于深度图像的绘制是三维视频系统中的一项关键技术。然而，深度编码失真会对后续虚拟视点绘制的性能产生极大的影响。提出了一种面向虚拟视点绘制的深度图像后处理方法，利用基于置信区间的交点规则确定深度图像中每个像素点的滤波窗口，利用最小均方逼近多项式技术估计每个像素点的深度值，再利用加权模式滤波器对估计的深度图像进行滤波处理，得到深度滤波图像。实验结果表明，该方法能够精确地恢复出真实深度信息，提高虚拟视点图像的绘制性能。

关键词: 深度图像后处理, 置信区间的交点, 最小均方逼近多项式技术, 加权模式滤波器