README.md

---
license: mit
pipeline_tag: image-to-3d
---

# StructSplat: Generalizable 3D Gaussian Splatting from Uncalibrated Sparse Views

This repository contains the pretrained weights for **StructSplat**, a feed-forward and generalizable 3D Gaussian reconstruction framework that operates directly on uncalibrated images without requiring camera parameters.

* **Paper:** [StructSplat: Generalizable 3D Gaussian Splatting from Uncalibrated Sparse Views](https://huggingface.co/papers/2606.28321)
* **Project Page:** [https://structsplat.github.io](https://structsplat.github.io)
* **Code:** [https://github.com/J-C-Zhao/StructSplat](https://github.com/J-C-Zhao/StructSplat)

## Installation & Evaluation

To set up the environment and run training or evaluation, please refer to the instructions in the [GitHub Repository](https://github.com/J-C-Zhao/StructSplat).

### Setup Environment

```bash
conda create -n structsplat python=3.10.19
conda activate structsplat
pip install torch==2.4.0 torchvision==0.19.0 -i https://download.pytorch.org/whl/cu118
pip install -r requirements.txt
```

### Evaluation

Run the following command to evaluate the model:

```bash
python evaluation.py -c config/dl3dv.yaml
```

## Citation

If you find this work useful, please cite the paper:

```bibtex
@inproceedings{zhao2026structsplat,
  title={StructSplat: Generalizable 3D Gaussian Splatting from Uncalibrated Sparse Views},
  author={Zhao, Jia-Chen and Chen, Beiqi and Chen, Xinyang and Wang, Guangcong and Nie, Liqing},
  booktitle={European Conference on Computer Vision},
  year={2026}
}
```