Video Personalization and Caching for Resource Constrained Environments

Video playback in mobile devices is becoming increasingly popular. Since mobile devices are typically resource constrained, various video personalization strategies are used to provide personalized video content that is most relevant to the client’s request. In this paper, we propose a novel video personalization server and cache architecture, which uses automatic video segmentation and video indexing based on semantic video content. A novel cache design and a novel cache replacement algorithm, specifically suited for caching personalized video files are proposed. The proposed cache replacement algorithm is shown to significantly outperform several state-of-the-art cache replacement policies.