Frustum-PointPillars: A Multi-Stage Approach for 3D Object Detection using RGB Camera and LiDAR

Accurate 3D object detection is a key part of the perception module for autonomous vehicles. A better understanding of the objects in 3D facilitates better decision-making and path planning. RGB Cameras and LiDAR are the most commonly used sensors in autonomous vehicles for environment perception. Many approaches have shown promising results for 2D detection with RGB Images, but efficiently localizing small objects like pedestrians in the 3D point cloud of large scenes has remained a challenging area of research. We propose a novel method, Frustum-PointPillars, for 3D object detection using LiDAR data. Instead of solely relying on point cloud features, we leverage the mature field of 2D object detection to reduce the search space in the 3D space. Then, we use the Pillar Feature Encoding network for object localization in the reduced point cloud. We also propose a novel approach for masking point clouds to further improve the localization of objects. We train our network on the KITTI dataset and perform experiments to show the effectiveness of our network. On the KITTI test set our method outperforms other multi-sensor SOTA approaches for 3D pedestrian localization (Bird’s Eye View) while achieving a significantly faster runtime of 14 Hz.

PDF Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Object Localization KITTI Pedestrian Easy Frustrum-PointPillars AP 60.98 % # 1
Birds Eye View Object Detection KITTI Pedestrians Easy Frustrum-PointPillars AP 60.98 % # 2
3D Object Detection KITTI Pedestrians Easy Frustrum-PointPillars AP 51.22 % # 5
Birds Eye View Object Detection KITTI Pedestrians Hard Frustrum-PointPillars AP 48.30 % # 1
3D Object Detection KITTI Pedestrians Hard Frustrum-PointPillars AP 39.28 % # 7
Object Localization KITTI Pedestrians Hard Frustrum-PointPillars AP 48.30 % # 1
Object Localization KITTI Pedestrians Moderate Frustrum-PointPillars AP 52.23 % # 1
3D Object Detection KITTI Pedestrians Moderate Frustrum-PointPillars AP 42.89 % # 7
Birds Eye View Object Detection KITTI Pedestrians Moderate Frustrum-PointPillars AP 52.23 % # 1

Methods