Multi-View RGB-D Synchronized Video Acquisition and Temporally Coherent 3D Animation Reconstruction Using Multiple Kinects

Abstract

This chapter introduces a system for acquiring synchronized multi-view color and depth (RGB-D) video data using multiple off-the-shelf Microsoft Kinect and methods for reconstructing temporally coherent 3D animation from the multi-view RGB-D video data. The acquisition system is very cost-effective and provides a complete software-based synchronization of the camera system. It is shown that the data acquired by this framework can be registered in a global coordinate system and then can be used to reconstruct the 360-degree 3D animation of a dynamic scene. In addition, a number of algorithms to reconstruct a temporally-coherent representation of a 3D animation without using any template model or a-prior assumption about the underlying surface are also presented. It is shown that despite some limitations imposed by the hardware for the synchronous acquisition of the data, a reasonably accurate reconstruction of the animated 3D geometry can be obtained that can be used in a number of applications.