Skip to content
/ TurboSL Public
forked from parsa-spm/TurboSL

Official Code Release for TurboSL, CVPR2024

Notifications You must be signed in to change notification settings

guge2/TurboSL

 
 

Repository files navigation

TurboSL: 基于神经逆结构光的密集、精确、快速3D重建

安装

本项目基于以下源代码构建:

  1. 按照上述仓库的步骤安装PyTorch和tiny-cuda-nn。

  2. 使用以下命令安装依赖:

pip install -r code/requirements.txt

数据集

data目录中提供了结构光图像的示例数据集,包括原始和去畸变图像(data/images)、投影图案(data/patterns)、标定参数和物体掩码。

训练

code/configs/neus-sl-geoalbamb-capture-blur.yaml中配置训练参数,然后运行code/launch.py

训练输出将保存在code/exp/[experiment_id]目录下。

参考训练结果见code/train_david目录。

引用

@inproceedings{mirdehghan2024turbosl,
  title={Turbosl: Dense accurate and fast 3d by neural inverse structured light},
  author={Mirdehghan, Parsa and Wu, Maxx and Chen, Wenzheng and Lindell, David B and Kutulakos, Kiriakos N},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={25067--25076},
  year={2024}
}

致谢

感谢Yuanchen Guo提供的Instant-NGP神经表面重建实现。

About

Official Code Release for TurboSL, CVPR2024

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 90.6%
  • Jupyter Notebook 9.4%