This paper proposes a real-time RGB-D (red-green-blue depth) 3D SLAM (simultaneous localization and mapping) with GPU (graphic processing unit) acceleration. RGB-D data contains 2D image and per-pixel depth information. 6-DOF (degree-of-freedom) visua...
This paper proposes a real-time RGB-D (red-green-blue depth) 3D SLAM (simultaneous localization and mapping) with GPU (graphic processing unit) acceleration. RGB-D data contains 2D image and per-pixel depth information. 6-DOF (degree-of-freedom) visual odometry is obtained through the 3D-RANSAC (three-dimensional random sample consensus) algorithm with 2D image features and depth data. For speed up extraction of features, parallel computation is performed with GPU acceleration. After a feature man-ager detects loop closure, a graph-based SLAM algorithm optimizes full trajectory of the sensor.