Plenary talk II: Big video data analytics: Research challenges and trends

Abstract

In this talk, the challenges and research trends for processing big video data will be addressed. The increase in number of cameras feeds and video data has created a huge video data, where ordinary processing is no longer sufficient. On the other hand, every day, more cameras are installed for different purposes. The reduction in camera prices and the introduction of different new concepts and technologies, such as deep learning and HEVC in addition to the increased performance of GPU and network connectivity have participated in accelerating the huge increase of video data. Many emerging applications in surveillance, traffic management, retail, autonomous driving, medical analysis, etc. require processing of large number of cameras feeds in a relatively short time. In order to do so, big-video data processing requirements needed to be understood and the existing tools should be modified to handle video data more efficiently. The talk will address the major bottle necks and challenges for processing video data using current big data tools. The talk will address also current research trends that aim at improving the efficiency of big-video-data processing.