FreeSplatter: Pose-free Gaussian Splatting for Sparse-view 3D Reconstruction

ICCV 2025

This repo is the official implementation of FreeSplatter, a feed-forward framework capable of generating high-quality 3D Gaussians from uncalibrated sparse-view images and recovering their camera parameters in mere seconds.

https://github.com/user-attachments/assets/0c73b693-9428-46bd-843c-132434b9686f

⚙️ Installation

We recommend using Python>=3.10, PyTorch>=2.1.0, and CUDA>=12.1.

conda create --name freesplatter python=3.10
conda activate freesplatter
pip install -U pip

# Install PyTorch and xformers
# You may need to install another xformers version if you use a different PyTorch version
pip install torch==2.4.0 torchvision==0.19.0 --index-url https://download.pytorch.org/whl/cu121
pip install xformers==0.0.27.post2

# Install other requirements
pip install -r requirements.txt

🤖 Pretrained Models

We provide the following pretrained models:

💫 Inference

We recommand to start a gradio demo in your local machine, simply run:

python app.py

⚖️ License

FreeSplatter's code and models are licensed under the Apache 2.0 License with additional restrictions to comply with Tencent's open-source policies. Besides, the libraries Hunyuan3D-1 and BRIAAI RMBG-2.0 have their own non-commercial licenses.

:books: Citation

If you find our work useful for your research or applications, please cite using this BibTeX:

@article{xu2024freesplatter,
  title={FreeSplatter: Pose-free Gaussian Splatting for Sparse-view 3D Reconstruction},
  author={Xu, Jiale and Gao, Shenghua and Shan, Ying},
  journal={arXiv preprint arXiv:2412.09573},
  year={2024}
}