Quality test for 3D video streaming system over WLAN

Abstract

The performance and the quality of 3D video streaming over WLAN networks is influenced by different parameters such as network transmission and the encoding of the 3D video stream. In this paper, we analyze a proposed live 3D video streaming system that transmits a 3D live video through a wireless LAN network (WLAN). The live 3D video is captured from the Kinect RGB-D sensor, then a 3D point cloud is being generated and processed. Several filters are applied during preprocessing stage which handles the spatial redundancy, then before streaming an Octree compression is used to further reduces the temporal redundancy. In this paper a color based conditional filter is designed to reduce color information for each frame. The filter condition is defined using the mean and the standard deviation of the occurrences in the color histogram. The quality of the whole system which includes filtering, compression and streaming is evaluated using several measurements: compression ratio, total bytes per points, peak signal to noise ratio (PSNR) and Similarity Structure index (SSIM). The experimental results shows that the proposed system have a best PSNR of 14dB and SSIM 81% with compression ratio of 10.69% while the best compression ratio is about 12% with PSNR 12.05dB of and SSIM 78.64%.