High definition H.264/AVC subjective video database for evaluating the influence of slice losses on quality perception

Abstract

Prior to the construction or validation of objective video quality metrics, ground-truth data must be collected by means of a subjective video database. This database consists of (impaired) video sequences and corresponding subjective quality ratings. However, creating this subjective database is a timeconsuming and expensive task. There is an ongoing effort towards publishing such subjective video databases into the public domain. This facilitates the development of new objective quality metrics. In this paper, we present a new subjective video database consisting of impaired High Definition H.264/AVC encoded video sequences and associated quality ratings gathered from a subjective experiment. This database can be used freely to determine impairment visibility or estimate overall quality of a video in the case of lost slices due to network impairments.Index Terms-Video quality assessment, Subjective video quality, Objective video quality metric, Quality of Experience, Quality of Service

XLR (piXel Loss Rate): A Lightweight Indicator to Measure Video QoE in IP Networks

“…C. Pixel Loss Rate as Key Quality Indicator 1) Experimental Design: To understand whether XLR is a reasonable KQI for video packet losses, we have evaluated its performance as FR metric for video quality, comparing it with other FR pixel-based metrics using three well-known widelyused databases: (i) the VQEG-HDTV Database, created, validated and used by many labs worldwide during the HDTV project [42] (specifically, we have used the subset VQEG-HD2, made publicly available by IRCyNN [44]); (ii) the LIVE Mobile VQA Database generated by the University of Texas at Austin [45]; and (iii) the AVC HD Subjective Video (AHSV) Database developed by the University of Ghent [46]. They are briefly described next.…”

Low latency 3D image streaming in medicine

“…One of the technologies offering a solution could be network streaming, with the help of which we would be able to transmit the data from a central computer to the users so that the users would not have to invest in expensive devices [5][6][7][8]. If the user wants to see the digitized medical samples processed by the server, he would be able to do so on his own, everyday computer.…”