Intelligent Camera Control for Graphical Environments

Too often in the field of computer graphics, practitioners have been more concerned with the question of how to move a camera rather than  why  to move it. This thesis addresses the core question of  why  the camera is being placed and moved and uses answers to that question to provide a more convenient, more intelligent method for controlling virtual cameras in computer graphics. After discussing the general sorts of activities to be performed in graphical environments, this thesis then contains a derivation of some camera primitives that are required, and examines how they can be incorporated into different interfaces. A single, consistent, underlying framework for camera control across many different domains has been posited and formulated in terms of constrained optimization. Examples from different application domains demonstrate a variety of interface styles that have all been implemented on top of the underlying framework. Evaluations for each application are also given.