Learned pointcloud representations do not generalize well with an increase in distance to the sensor. For example, at a range greater than 60 meters, the sparsity of lidar pointclouds reaches a point where even humans cannot discern object shapes from each other. However, this distance should not be considered very far for fast-moving vehicles: a vehicle can traverse 60 meters in under two seconds while moving at 70 mph. For safe and robust driving automation, acute 3D object detection at these ranges is indispensable. Against this backdrop, we introduce faraway-frustum: a novel fusion strategy for detecting faraway objects. The main strategy is to depend solely on the 2D vision sensor for recognizing and classifying an object, as object shape does not change drastically with an increase in depth, and use pointcloud data for object localization in the 3D space for faraway objects. For closer objects, we use learned pointcloud representations instead, following state-of-the-art practices. This strategy alleviates the main shortcoming of object detection with learned pointcloud representations. Experiments on the KITTI dataset demonstrate that our method outperforms state-of-the-art methods by a considerable margin for faraway object detection in bird's-eye view and 3D. Our code is open-source and publicly available: https://github.com/dongfang-steven-yang/faraway-frustum.
Faraway-Frustum: Dealing with Lidar Sparsity for 3D Object Detection using Fusion
19.09.2021
3070509 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
TEMP-FRUSTUM NET: 3D OBJECT DETECTION WITH TEMPORAL FUSION
British Library Conference Proceedings | 2021
|HIGH DIMENSIONAL FRUSTUM POINTNET FOR 3D OBJECT DETECTION FROM CAMERA, LIDAR, AND RADAR
British Library Conference Proceedings | 2020
|