Radar-to-Lidar: Heterogeneous Place Recognition via Joint Learning
radar-to-lidar-place-recognition
This page is the implementation of a pre-print, implemented by PyTorch.
If you have some questions on this project, please feel free to contact Huan Yin [email protected] .
Method
Data
The files in matlab/RobotCar_data and matlab/MulRan_data can help you generate scancontext of radar and lidar submaps. Also, the generation of lidar submaps is included.
Training
The train_disco_lidar_quad.py is used for training lidar-to-lidar DiSCO.
The train_disco_radar_quad.py is used for training radar-to-radar DiSCO.
The train_joint_radar_lidar.py is used for training L2L, R2R and R2L jointly based on DiSCO implementation.
The trained models are listed in the trained_models respectively.
Inference
Please use the files in inference folder.
Evaluation
In addition, the matlab/[email protected] contains the files to