MANet: End-to-End Learning for Point Cloud Based on Robust Pointpillar and Multiattention

Figure 8

Results of 3D detection on the KITTI test set. For better visualization, the 3D boxes detected using lidar are projected onto images from the left camera.