Multi-View 3D Object Detection Network for Autonomous Driving
About
This paper aims at high-accuracy 3D object detection in autonomous driving scenario. We propose Multi-View 3D networks (MV3D), a sensory-fusion framework that takes both LIDAR point cloud and RGB images as input and predicts oriented 3D bounding boxes. We encode the sparse 3D point cloud with a compact multi-view representation. The network is composed of two subnetworks: one for 3D object proposal generation and another for multi-view feature fusion. The proposal network generates 3D candidate boxes efficiently from the bird's eye view representation of 3D point cloud. We design a deep fusion scheme to combine region-wise features from multiple views and enable interactions between intermediate layers of different paths. Experiments on the challenging KITTI benchmark show that our approach outperforms the state-of-the-art by around 25% and 30% AP on the tasks of 3D localization and 3D detection. In addition, for 2D detection, our approach obtains 10.3% higher AP than the state-of-the-art on the hard data among the LIDAR-based methods.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| 3D Object Detection | KITTI car (test) | AP3D (Easy)74.97 | 195 | |
| 3D Object Detection | Waymo Open Dataset (val) | -- | 175 | |
| 3D Object Detection | KITTI (val) | AP3D (Moderate)62.68 | 85 | |
| 3D Object Detection | KITTI (test) | AP_3D (Easy)71.1 | 83 | |
| Bird's Eye View Detection | KITTI Car class official (test) | AP (Easy)86.62 | 62 | |
| 3D Object Detection | KITTI (test) | 3D AP Easy74.97 | 61 | |
| 3D Object Detection | KITTI (test) | AP_3D Car (Easy)74.97 | 60 | |
| Bird's Eye View Object Detection (Car) | KITTI (test) | APBEV (Easy) @IoU=0.786.62 | 59 | |
| 3D Object Detection | KITTI (val) | -- | 57 | |
| 3D Object Detection | KITTI cars (val) | AP Easy71.29 | 48 |