Analysis of Spatial and Temporal Information of DASH Dataset

In order to provide the end user with the best possible quality of requested video content, standard called MPEG Dynamic Adaptive Streaming over HTTP was developed. It defines a way of notifying the user that a variety of transmission streams of different qualities exist together with the information needed to select the appropriate transmission path. In order to analyze the quality of videos provided to users, Quality of Service and Quality of Experience parameters should be taken into account. In order to improve content segmentation as well as the selection of the best adaptation scenario, it is necessary to analyze the content itself. In this paper temporal and spatial activity are analyzed for three video resolutions and several bit rates. According to PSNR and SSIM results, spatial activity has high influence on the video quality and can be useful as a parameter in DASH dataset preparation in order to improve adaptation algorithms and thus users Quality of Experience.