Learning Moving-Object Tracking with FMCW LiDAR

In this paper, we propose a learning-based moving-object tracking method utilizing the newly developed LiDAR sensor, Frequency Modulated Continuous Wave (FMCW) LiDAR. Compared with most existing commercial LiDAR sensors, FMCW LiDAR can provide additional Doppler velocity information to each 3D point of the point clouds. Benefiting from this, we can generate instance labels as ground truth in a semi-automatic manner. Given the labels, we propose a contrastive learning framework, which pulls together the features from the same instance in embedding space and pushes apart the features from different instances, to improve the tracking quality. Extensive experiments are conducted on the recorded driving data, and the results show that our method outperforms the baseline methods by a large margin.

Authors

• Yi Gu
• Hongzhi Cheng
• Kafeng Wang
• Dejing Dou
• ChengZhong Xu 0001
• Hui Kong 0001

Keywords

• Point cloud compression
• Laser radar
• Three-dimensional displays
• Frequency modulation
• Annotations
• Semantics
• Pipelines
• Frequency Modulated Continuous Wave
• Point Cloud
• Latent Space
• Learning-based Methods
• Baseline Methods
• Self-supervised Learning
• Velocity Information
• LiDAR Sensor
• Deep Learning
• Supervised Learning
• Window Size
• Bounding Box
• Segmentation Accuracy
• Distance Threshold
• Clusters Of Points
• Density Parameters
• Doppler Shift
• Nearest Neighbor Search
• Semantic Labels
• Multi-object Tracking
• Semi-automated Process
• Cluster Position
• Dynamic Point
• Ground Points
• Density-based Clustering
• Gaussian Assumption
• Symmetric Function
• Bounding Box Annotations
• Labeled Data