Awesome
InAction (ECCV 2022)
Implementation of the ECCV work "InAction" for interpretable driving behavior prediction.
Data
Dependencies
- Python 3.6
- Pytorch 1.1
Training
Evaluation
- Single-label Benchmarks (PSI)
Driving action prediction: Accuracy & Class-wise average accuracy
Explanation: F1 score & Class-wise average F1 score
- Multiple-label Benchmarks (BDD-OIA)
Driving action prediction: F1 score & Class-wise average F1 score
Explanation: F1 score & Class-wise average F1 score
Citation
If you think this work is interesting, please cite:
Taotao Jing1, Haifeng Xia1, Renran Tian2, Haoran Ding2, Xiao Luo2, Joshua
Domeyer3, Rini Sherony3, and Zhengming Ding1
@InProceedings{Jing_2021_InAction,
author = {Jing, Taotao and Xia, Haifeng and Tian, Renran and Ding, Haoran and Luo, Xiao and Domeyer, Joshua and Sherony, Rini and Ding, Zhengming},
title = {InAction: Interpretable Action Decision Making for Autonomous Driving},
booktitle = {European conference on computer vision},
year = {2022}
}
Contact
If you have any questions about this work, feel free to contact