Robust real-time stereo matching system for indoor environments

Stereo matching is a traditional method used for obtaining three-dimensional depth information and has been studied for decades. However, it is still difficult to apply stereo matching algorithms to practical devices because of certain real-time issues and performance issues due to untextured regions, radial distortion, etc. Therefore, recent applications use active methods such as structured light (Kinect). However, applications using Kinect cannot accurately calculate the disparity for thin objects at distances of more than 3 m, because they use low-resolution images and structured patterns having points separated by large distances. In this paper, we propose a real-time stereo matching system to remedy the disadvantages of active and passive stereo vision. Further, we present a video that shows a qualitative comparison between Kinect 1, Kinect 2, and the proposed system, including point cloud results.