An open API service indexing awesome lists of open source software.

https://github.com/atticuszeller/gsplatloc

😎 GsplatLoc 🎯: Ultra-Precise Pose Optimization via 3D Gaussian Reprojection 🌟
https://github.com/atticuszeller/gsplatloc

3d-reconstruction 6dof camerapose-estimate computer-vision deeplearning gaussian-splatting gsplat localization python pytorch slam

Last synced: about 2 months ago
JSON representation

😎 GsplatLoc 🎯: Ultra-Precise Pose Optimization via 3D Gaussian Reprojection 🌟

Awesome Lists containing this project

README

        


GSplatLoc: Ultra-Precise Camera Localization via 3D Gaussian Splatting



Atticus Zeller





## ⚙️ Setting Things Up

> CUDA 12.1 [download](https://developer.nvidia.com/cuda-12-1-0-download-archive?target_os=Linux&target_arch=x86_64&Distribution=Ubuntu&target_version=22.04&target_type=deb_local)

Install [UV](https://docs.astral.sh/uv/getting-started/installation/#standalone-installer):

```bash
curl -LsSf https://astral.sh/uv/install.sh | sh
```

Then setup environment from the `pyproject.toml` :

```bash
uv sync --all-extras --all-groups --dev
source .venv/bin/activate
```

## 🔨 Running GsplatLoc

Downloading the Data:

```bash
bash scripts/download_datasets.sh
```

Initialize the Weights & Biases logging:
1. use wandb login with api key https://wandb.ai/authorize, then `wandb login --relogin`
2. update `.env` with your wandb content.

Reproducing Results:

```bash
export MPLBACKEND=Agg
bash scripts/run_eval.sh
```

## 📌 Citation

If you find our paper and code useful, please cite us:

```bib
@misc{zeller2024gsplatlocultraprecisecameralocalization,
title={GSplatLoc: Ultra-Precise Camera Localization via 3D Gaussian Splatting},
author={Atticus J. Zeller},
year={2024},
eprint={2412.20056},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2412.20056},
}
```