![]() In our VIO method, the depth information is introduced to improve the accuracy of pose estimation, and FAST features are used for faster tracking. The CDSFusion is the first system integrating RGBD-based Visual-Inertial Odometry (VIO), semantic segmentation and 3D reconstruction in real-time on a CPU. To solve the problem, an indoor dense semantic Simultaneous Localization and Mapping (SLAM) method using CPU computing is proposed in this paper, named CDSFusion. Existing methods rely on GPU, and it is difficult to achieve real-time semantic reconstruction on CPU. However, high-performance computing is generally not available on most UAVs, so a lightweight real-time semantic reconstruction method is necessary. The semantic reconstruction of the scene is a truly functional understanding of the environment. Unmanned Aerial Vehicles (UAVs) require the ability to robustly perceive surrounding scenes for autonomous navigation.
0 Comments
Leave a Reply. |