---
license: apache-2.0
pipeline_tag: image-to-3d
tags:
- Feed-forward
- 3D Gaussian Splatting
- Camera Pose Estimation
- Multi-view Images
---

<h2 align=center font-weight:100> <strong><i>VicaSplat</i></strong>: A Single Run is All You Need for 3D Gaussian Splatting and Camera Estimation from Unposed Video Frames</h2>

<h5 align="center">

[![arXiv](https://img.shields.io/badge/ArXiv-2503.10286-b31b1b.svg?logo=arXiv)](https://arxiv.org/abs/2503.10286)
[![Project page](https://img.shields.io/badge/Project-Page-brightgreen)](https://lizhiqi49.github.io/VicaSplat/)
[![HuggingFace](https://img.shields.io/badge/HuggingFace-Model-blue?style=flat&logo=HuggingFace)](https://huggingface.co/lzq49/VicaSplat)

</h5>

<p align="center">
  <img src="teaser.jpg" width="80%">
</p>

Official implementation of **VicaSplat** - feed-forward 3D Gaussians reconstruction and camera pose estimation from unposed video frames in ~0.1s!

## Citation

```
@article{li2025vicasplat,
      title   = {VicaSplat: A Single Run is All You Need for 3D Gaussian Splatting and Camera Estimation from Unposed Video Frames},
      author  = {Zhiqi Li and Chengrui Dong and Yiming Chen and Zhangchi Huang and Peidong Liu},
      journal = {arXiv preprint arXiv:2503.10286},
      year    = {2025}
    }
```