Complete processing chain for 3D video generation using Kinect sensor

Abstract

The multiview-video-plus-depth (MVD) format selected for 3D video standardization describes 3D scene by video and associated depth and enables generation of virtual views through the depth-image-based rendering (DIBR) process. The Kinect™ sensor is equipped with a red-green-blue (RGB) camera and a depth sensor and it is suitable to capture synchronous video and depth streams. However, these sensors do not produce data with pixel-wise correspondence that is required by the DIBR process.