This is the official repository of our submitted work "NeuroIV: Neuromorphic Vision Meets Intelligent Vehicle towards Safe Driving with a New Database and Baseline Evaluations". If you use any of them, please cite:
@ARTICLE{9234108,
author={Chen, Guang and Wang, Fa and Li, Weijun and Hong, Lin and Conradt, Jörg and Chen, Jieneng and Zhang, Zhenyan and Lu, Yiwen and Knoll, Alois},
journal={IEEE Transactions on Intelligent Transportation Systems},
title={NeuroIV: Neuromorphic Vision Meets Intelligent Vehicle Towards Safe Driving With a New Database and Baseline Evaluations},
year={2022},
volume={23},
number={2},
pages={1171-1183},
doi={10.1109/TITS.2020.3022921}}
Neuromorphic vision sensors such as the Dynamic and Active-pixel Vision Sensor (DAVIS) using silicon retina are inspired by biological vision, they generate streams of asynchronous events to indicate local log-intensity brightness changes. Their properties of high temporal resolution, low-bandwidth, lightweight computation, and low-latency make them a good fit for many applications of motion perception in the intelligent vehicle. However, as a younger and smaller research field compared to classical computer vision, neuromorphic vision is rarely connected with the intelligent vehicle. For this purpose, we release a novel dataset, NeuroIV, which comprises of three datasets recorded with DAVIS sensors and Depth camera.
DAVIS346redColor
Realsense D435i
| Datasets | Classes |
|---|---|
| Driver Drowsiness | 1. Normal driving 2. Drowsiness driving |
| Driver Gaze-Zone | 1. To the left 2. To left stack 3. Switch downward 4. To center stack 5. To the right |
| Driver Hand-Gesture | 1. Expand 2. MoveDown 3. MoveLeft 4. MoveRight 5. MoveUp 6. No 7. OK 8. OneTap 9. Pinch 10. PushDown 11. PushLeft 12. PushRight 13. PushUp 14. V 15. Wiper 16. X |
Download link: https://pan.baidu.com/s/10AQ-0tMoPW2XnL0LpDlVUQ and the code: 42uj.