WTEFNet: Real-Time Low-Light Object Detection for Advanced Driver Assistance Systems

Hao Wu, Junzhou Chen, Ronghui Zhang, Nengchao Lyu, Hongyu Hu, Yanyong Guo, and Tony Z. Qiu

Paper Download

• The download link for the GSN dataset is below:

Google Drive	Download
Baidu Cloud	Download (Extraction code: uvbm)

• The file structure of the downloaded dataset is as follows.

output
├── images
│   ├── train
│   ├── val
│   │── test
├── labels
│   ├── train
│   ├── val
│   │── test

👐 Hands-On Guide

Step 1 — Requirements

To install requirements:

pip install -e .

Step 2 — Prepare the configuration file

model: ./ultralytics/cfg/models/wtefnet.yaml
dataset: ./ultralytics/data/mydataset.yaml

Step 3 - Train

cd ultralytics
python train.py

Step 4 - Test

python test.py

Citation

If you use WTEFNet or GSN dataset, please consider citing:

@article{wu2025wtefnet,
  title={WTEFNet: Real-Time Low-Light Object Detection for Advanced Driver-Assistance Systems},
  author={Wu, Hao and Chen, Junzhou and Zhang, Ronghui and Lyu, Nengchao and Hu, Hongyu and Guo, Yanyong and Qiu, Tony Z},
  journal={arXiv preprint arXiv:2505.23201},
  year={2025}
}