Faraway-Frustum: Dealing with Lidar Sparsity for 3D Object Detection using Fusion

by   Haolin Zhang, et al.

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 to 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 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 for recognizing object class, 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. 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 by a considerable margin for faraway object detection in bird's-eye-view and 3D.


page 1

page 2


BirdNet+: End-to-End 3D Object Detection in LiDAR Bird's Eye View

On-board 3D object detection in autonomous vehicles often relies on geom...

CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection

There have been significant advances in neural networks for both 3D obje...

Towards a Robust Sensor Fusion Step for 3D Object Detection on Corrupted Data

Multimodal sensor fusion methods for 3D object detection have been revol...

On the Role of Sensor Fusion for Object Detection in Future Vehicular Networks

Fully autonomous driving systems require fast detection and recognition ...

High-level camera-LiDAR fusion for 3D object detection with machine learning

This paper tackles the 3D object detection problem, which is of vital im...

AdaptiveShape: Solving Shape Variability for 3D Object Detection with Geometry Aware Anchor Distributions

3D object detection with point clouds and images plays an important role...

DOPS: Learning to Detect 3D Objects and Predict their 3D Shapes

We propose DOPS, a fast single-stage 3D object detection method for LIDA...

Please sign up or login with your details

Forgot password? Click here to reset