Multi-camera reconstruction and rendering for free-viewpoint video

While virtual environments in interactive entertainment become more and more lifelike and sophisticated, traditional media like television and video have not yet embraced the new possibilities provided by the rapidly advancing processing power. In particular, they remain as non-interactive as ever, and do not allow the viewer to change the camera perspective to his liking. The goal of this work is to advance in this direction, and provide essential ingredients for a free-viewpoint video system, where the viewpoint can be chosen interactively during playback. Knowledge of scene geometry is required to synthesize novel views. Therefore, we describe 3D reconstruction methods for two distinct kinds of camera setups. The first one is depth reconstruction for camera arrays with parallel optical axes, the second one surface reconstruction, in the case that the cameras are distributed around the scene. Another vital part of a 3D video system is the interactive rendering from different viewpoints, which has to perform in real-time. We cover this topic in the last part of this thesis.