Skip to content

danaiefst/GaitTracking

Repository files navigation

GaitTracking

This is the repository for the code of paper Deep Leg Tracking by Detection and Gait Analysis in 2D Range Data for Intelligent Robotic Assistants.

The database used for training the network can be found here.

On main branch exists the Leg Tracking network training code On gait branch exists the Gait Analysis network training code

File decription:

  • model.pt, best_model.pt: PyTorch pretrained models
  • data_handler.py: Custom DataLoader that turns laserpoints into occupancy grids and also loads center and gait state annotations
  • check_batches.py: An example usage of the data_handler.py for visual representation of the data
  • tracking_nn.py: The network architecture
  • train.py: The training and validation code
  • test.py: The final testing code

File paths should be changed in order for the code to run.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages