A Theoretical Framework for End-to-End Video Quality Prediction of MPEG-based Sequences

— This paper presents a novel theoretical framework for end-to-end video quality prediction of MPEG-based video sequences. The proposed framework encloses two discrete models: i) A model for predicting the video quality of an encoded signal at a pre-encoding stage and ii) A model for mapping QoSsensitive network parameters (i.e. packet loss) to video quality degradation. The efficiency of both the discrete models is experimentally validated, proving by this way the accuracy of the proposed framework. Keywords-Video Quality; Packet Loss;MPEG;H.264

• Type: Conference
• Year: 2007
• Where: ICNS
• Authors: Harilaos Koumaras, Anastasios Kourtis, Cheng-Han Lin, Ce-Kuen Shieh
• Keywords: Computer Networks, Discrete Models, ICNS 2007, Video Quality, Video Quality Prediction