Intelligent video production system for recording presentations

摘要

In this paper we introduce an intelligent system for video or multimedia contents production. First, we categorized targets and purpose of shooting, and discuss the cameraworks appropriate for them. Then, we propose camera control algorithms to realize the cameraworks. Moreover, weintroduce video tagging by recognizing a speaker's behaviors, such as deictic movements. In the experiments on the prototype system, our algorithms showed good performance.