Perceptual video quality measurement for streaming video over mobile networks

Over the last decade there has been tremendous progress in video compression and data communication technologies that provide the basis for video streaming services. This has led to rapid deployment of mobile devices capable of capturing and displaying images and video which in turn provides new technical challenges and commercial opportunities for video streaming technologies. This emerging trend in providing multimedia services like streaming video, video conferencing and games over mobile networks has lead to the study of visual quality of the transmitted video sequences. The quality of all these services is based upon the Quality of Experience (QoE) of the user. This thesis focuses on methods for measuring video quality objectively to identify QoE as perceived by a customer when viewing streaming video transmissions over Internet. The results of the thesis will give an understanding of the factors effecting quality of mobile video transmissions and the information can be used for providing better video quality. If we can actually identify the amount of distortions that are actually able to perceive by the user then we can estimate the quality of the video sequence based on those details. Based on this idea and an understanding of human visual system, we implemented a simple but effective video quality pipeline for evaluating the perceptual video quality.