一种利用背景信息的虚拟视点图像后处理方法

Virtual Viewpoint Image Post-processing Method Using Background Information

基于深度图像的虚拟视点绘制(Depth Image Based Rendering,DIBR)在双视图像融合后存在重叠、伪影和空洞等问题,为提高融合后图像质量,提出一种利用背景信息的虚拟视点图像后处理方法.首先,在图像融合时引入深度信息解决像素重叠问题.其次,根据视点移动方向定位左、右虚拟视图伪影区域,利用互补视点的背景像素擦除伪影.最后,利用不同视点参考图像信息之间的冗余性将左、右参考视点进行融合并消除前景像素,再进行正向映射得到虚拟视点背景图像填充双视融合后剩余的空洞.与基于图像分割的虚拟视点绘制算法相比PSNR值提高了0.6357dB,实验结果表明,该算法能有效提高绘制图像质量.

Depth Image Based Rendering(DIBR)has some problems,such as overlap,artifacts and voids,after the fusion of two views.In order to improve the quality of the fused Image,a virtual viewpoint image post-processing method using background information was proposed.Firstly,depth information is introduced to solve the problem of pixel overlap in image fusion.Secondly,the artifact regions of the left and right virtual views are positioned according to the viewpoint movement direction,and the artifacts are erased by the background pixels of the complementary viewpoints.Finally,the redundancy between the reference images of different viewpoints is used to fuse the left and right reference points and eliminate the foreground pixels,and then forward mapping is carried out to get the background image of the virtual viewpoint to fill the remaining void after the fusion of the two views.Compared with the virtual viewpoint rendering algorithm based on image segmentation,the PSNR value is improved by 0.6357dB.Experimental results show that the proposed algorithm can effectively improve the quality of rendering images.