الفهرس 
3D semantic segmentation using camera-lidar sensor fusion for autonomous driving
English SummaryOnly 14 pages are availabe for public view
 |  | from | 100 |  |  |
| | | from | 100 | | |
Abstract
This research improves the environment sensing in control systems through multi-sensor fusion. We are fusing Camera raw image data with LiDAR Point cloud resulting from having a colored point cloud, represented in a spherical grid map representation. We used the fused data in early and mid-level fusion algorithms. We evaluate our algorithms on the KITTI dataset, which provides semantic segmentation for different classes such as Cars, Pedestrians, and cyclists. We evaluated our work on two states of art architectures, namely Squeeze Seg and Point Seg, resulting in improving the mIOU score to 10% in both cases relative to the LiDAR only baseline