python intensity_analyser.py
 

Extracts the data and calibrates then for training, generating ground truth (Intensity ranges of classes), plotting etc.

 python /utils/GT_corrector.py
 

To correct the outliers in the ground truth and data generated from <intensity_analyser.py>

 python /alpha_predictor/alpha_model.py
 

Contains the ANN architecture of the predictor.

 python /alpha_predictor/train_alpha.py
 

Command to train the model. Please edit the paths to the root file of the dataset.

 python intensity_predictor.py
 

  python intensity_predictor_velodyne.py
 

  python ins2ref.py
 

Command to predict the classes of the LiDAR points. Reads .ply files and predicts for classes: grass, bush, trees, puddle, person.

 python /utils/make_video.py
 python /utils/data_counter.py
 

To generate movies from images.

If you find this work useful for your research, do cite us.

@InProceedings{10.1007/978-3-031-63596-0_54,
author="Viswanath, Kasi
and Jiang, Peng
and Sujit, P. B.
and Saripalli, Srikanth",
editor="Ang Jr, Marcelo H.
and Khatib, Oussama",
title="Off-Road LiDAR Intensity Based Semantic Segmentation",
booktitle="Experimental Robotics",
year="2024",
publisher="Springer Nature Switzerland",
address="Cham",
pages="608--617",
isbn="978-3-031-63596-0"
}