Skip to content

shulnak09/Uncertainty-Estimation-using-Bayesian-Approximation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 

Repository files navigation

Uncertainty-Estimation-using-Bayesian-Approximation

In this work, we estimate the posterior predictive distribution using Monte Carlo dropout during trajectory forecasting for pedestrians. Our probabilistic model using MC dropout outperforms previous deterministic methods in terms of perfromance metrics like ADE and FDE.

Sample Trajectoy prediction with uncertainty. As per literature, we observe 8 (3.2 secs) historical steps to predict 12 steps (4.8 secs) into future.

CNN_LSTM_ETH_UQ_do_0 2_and_FP_4 8_secs

cite Contains the tensorflow implementation of our [paper] (https://ieeexplore.ieee.org/document/9882968) on oogle colab. If you find this code helpful in your research, please consider citing our paper:

@ARTICLE{9882968, author={Nayak, Anshul and Eskandarian, Azim and Doerzaph, Zachary}, journal={IEEE Open Journal of Intelligent Transportation Systems}, title={Uncertainty Estimation of Pedestrian Future Trajectory Using Bayesian Approximation}, year={2022}, volume={3}, number={}, pages={617-630}, doi={10.1109/OJITS.2022.3205504}}

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published