Robust computer vision techniques for high-quality 3D modeling

In this paper, we present our recent sensor fusion approaches to obtain high-quality 3D information. We first discuss two fusion methods that combine geometric and photometric information. The first method, multiview photometric stereo, reconstructs the full 3D shape of a target object. The geometric and photometric information is efficiently fused by using a planar mesh representation. The second method performing shape-from shading with a Kinect sensor estimates the shape of an object under uncalibrated natural illumination. Since the method uses a single RGB-D input, it is capable of capturing the high quality shape details of a dynamic object under varying illumination. Subsequently, we summarize a calibration algorithm of a time of-flight (ToF) sensor and a camera fusion system with a 2.5D pattern. Lastly, we present a camera-laser sensor fusion system for the large-scale 3D reconstruction.