Performance Analysis of Large-scale Realtime Video Stream Configurations

University essay from Göteborgs universitet/Institutionen för data- och informationsteknik

Abstract: The rapid technology development has resulted insensors able to deliver very high-quality output. The high qualitycalls for efficient compression algorithms to handle the vastamount of data produced. This study aims to find a compressionalgorithm that delivers video streams of the highest possiblequality given the constraints real-time processing using the UserDatagram Protocol (UDP). This paper describes the experimentalapproach created to find such compression algorithm. Machinelearning in the form of Bayesian optimization was applied toevaluate and hence deduce the optimal encoder parameters foreach encoder and resolution in scope.WebM Project’s VP9 implementation proved to be the mostoptimal encoder in scope for all resolutions evaluated in the experimentbut the highest (QXGA - 2048x1536). For video streamsin QXGA, VP9 hardware accelerated by Intel’s QuickSync wasfound to perform best.

  AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)