README.md

Model Output Samples

This folder contains the output of state-of-the-art algorithms, ENCORE and PedFormer for trajectory and action prediction, intention estimation, and risk assessment tasks.

Files are formatted as <model_abbrev>_<task>_<dataset>.csv. To load the model output files use from utilities.utils import get_predictions. See scenario script for use-case example. The outputs can be used to replicate the results in the following papers:

Citation

If you use the model outputs for evaluation, please cite the corresponding paper(s).

For ENCORE trajectory prediction:

@InProceedings{Rasouli_2024_ICRA,
author = {Rasouli, Amir},
title = {A Novel Benchmarking Paradigm and a Scale- and Motion-Aware Model for Egocentric Pedestrian Trajectory Prediction},
booktitle = {International Conference on Robotics and Automation (ICRA)},
year = {2024}
}

For PedFormer behavior prediction and estimation:

@InProceedings{Rasouli_2024_IV,
author = {Rasouli, Amir and Kotseruba, Iuliia},
title = {Diving Deeper Into Pedestrian Behavior Understanding: Intention Estimation, Action Prediction, and Event Risk Assessment},
booktitle = {Intelligent Vehicles Symposium (IV)},
year = {2024}
}

@InProceedings{Rasouli_2023_ICRA,
author = {Rasouli, Amir and Kotseruba, Iuliia},
title = {PedFormer: Pedestrian Behavior Prediction via Cross-Modal Attention Modulation and Gated Multitask Learning},
booktitle = {International Conference on Robotics and Automation (ICRA)},
year = {2023}
}