duo / README.md
hzcar's picture
Update README.md
d4cdfaa verified
---
license: mit
size_categories:
- 10K<n<100K
---
# Dataset README
This repository provides all datasets and model checkpoints required for the paper:
[Adaptive Dual Uncertainty Optimization: Boosting Monocular 3D Object Detection under Test-Time Shifts 🔗](https://arxiv.org/abs/2508.20488)
*Zixuan Hu, Dongxiao Li, Xinzhu Ma, Shixiang Tang, Xiaotong Li, Wenhan Yang, Ling-Yu Duan*
(**ICCV 2025 Highlight**)
---
## Repository Structure
- **kitti_noise.zip**, **kitti_blur.zip**, **kitti_weather.zip**, and **kitti_digital.zip**
After extraction, together they form the complete **KITTI-C** dataset.
- **Nuscenes.zip**
Contains the splitted datasets across four different scenarios.
- **source_ckpt.zip**
Includes all necessary model checkpoints.
---
## Notes
⚠️ The corruption process in **KITTI-C** involves certain randomness, which may lead to slight fluctuations in detection performance.
If you use the dataset provided in this repository, please make sure to properly acknowledge it in your citation.
---
## Project Website
👉 [https://github.com/hzcar/DUO](https://github.com/hzcar/DUO)
## Citation
If our datasets and models are helpful in your research, please consider citing our paper:
```
@article{hu2025adaptive,
title={Adaptive Dual Uncertainty Optimization: Boosting Monocular 3D Object Detection under Test-Time Shifts},
author={Hu, Zixuan and Li, Dongxiao and Ma, Xinzhu and Tang, Shixiang and Li, Xiaotong and Yang, Wenhan and Duan, Ling-Yu},
journal={arXiv preprint arXiv:2508.20488},
year={2025}
}
```