Skip to Main Content
In this paper we describe an approach for 3D-TV Layered Depth Video (LDV) - Content creation using a capturing system of four CCD - Cameras and Time-Of-Flight-Sensor (ToF - Camera). We demonstrate a whole video production chain, from calibration of the camera rig, to generation of reliable depth maps for a single view of one of the CCD - Cameras, using only estimated depth provided by the ToF - Camera. We additionally show that we are able to generate proper occlusion layers for LDV - Content through a straight forward approach based on depth background extrapolation and backward texture mapping.
Date of Conference: 4-6 May 2009