TripleMixer: A 3D Point Cloud Denoising Model for Adverse Weather

This is the official repository of the Weather-KITTI and Weather-NuScenes dataset. For technical details, please refer to:

TripleMixer: A 3D Point Cloud Denoising Model for Adverse Weather
Xiongwei Zhao*, Congcong Wen*, Yang Wang, Haojie Bai, Wenhao Dou.

**[Paper] [Blog][Download] **

1. Dataset

(1) Overview

In this Work, we propose our synthetic adverse weather datasets, named Weather-KITTI and Weather-NuScenes, which are based on the SemanticKITTI and nuScenes-lidarseg datasets, respectively. These datasets cover three common adverse weather conditions: rain, fog, and snow and retain the original LiDAR acquisition information and provide point-level semantic labels for rain, fog, and snow.

(2) Dataset Statistics

2. Quantitative Result of Denoising Task

3. Training and Evaluation

Continuous Updates！

Citation

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

@misc{zhao2024triplemixer3dpointcloud,
      title={TripleMixer: A 3D Point Cloud Denoising Model for Adverse Weather}, 
      author={Xiongwei Zhao and Congcong Wen and Yang Wang and Haojie Bai and Wenhao Dou},
      year={2024},
      eprint={2408.13802},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2408.13802}, 
}

Updates

24/08/2024: Initial release and submitted to the Journal. The dataset will be open source soon!

License

The dataset is based on the SemanticKITTI dataset, provided under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 United States License (CC BY-NC-SA 3.0 US), and the nuScenes-lidarseg dataset, provided under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0). This dataset is provided under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0).

Grandzxw/TripleMixer