TPNSTA

Code and models for IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING paper: Temporal Pyramid Network with Spatial-Temporal Attention for Pedestrian Trajectory Prediction

Environment

• Python 3.8
• pytorch 1.11.0
• cuda 11.3
• Ubuntu 20.04
• RTX 3090
• Please refer to the "requirements.txt" file for more details.

Usage

To test the model, run:

scripts/evaluate_model_merge_standar_Offline_Add.py

To train the model, run:

scripts/train_multi_scale_merge_standar_OFFline_Ave_st_first.py

Citation

If you find this work useful in your research, please consider citing:

@ARTICLE{9373939,  author={Li, Yuanman and Liang, Rongqin and Wei, Wei and Wang, Wei and Zhou, Jiantao and Li, Xia},  journal={IEEE Transactions on Network Science and Engineering},   title={Temporal Pyramid Network with Spatial-Temporal Attention for Pedestrian Trajectory Prediction},   year={2021},  volume={},  number={},  pages={1-1},  doi={10.1109/TNSE.2021.3065019}}

Contact

If you encounter any issue when running the code, please feel free to reach us either by creating a new issue in the github or by emailing

• 1810262064@email.szu.edu.cn