Videography for Telepresentations

Our goal is to help automate the capture and broadcast of lectures to remote audiences. There are two components to the design of such systems. The first is the technology component that includes the hardware (e.g. video cameras) and associated software (e.g., computer vision software to track the lecturer). The second is the aesthetic component that embodies the rules and idioms that human videographers follow to make the video visually engaging. In this paper we present detailed video-production rules obtained from professional videographers as they critiqued a state-of-the-art lecture room automation system. To facilitate the construction of similar systems by other practitioners in the field, we present generalized rules for different lecture room environments, assuming different numbers and types of cameras. We further discuss areas of gap between what the professional videographers require and what is technologically feasible today.