Hugging Face's logo

g3splat
/
g3splat

Image-to-3D
PyTorch
English
3d-reconstruction
gaussian-splatting
novel-view-synthesis
depth-estimation
pose-estimation
computer-vision
3dgs
2dgs
feed-forward
self-supervised
Model card Files
xet
Community
g3splat
File size: 2,918 Bytes 
5cae0ec
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
 
---
license: mit
tags:
  - 3d-reconstruction
  - gaussian-splatting
  - novel-view-synthesis
  - depth-estimation
  - pose-estimation
  - computer-vision
  - pytorch
  - 3dgs
  - 2dgs
  - feed-forward
  - self-supervised
language:
  - en
pipeline_tag: image-to-3d
library_name: pytorch
arxiv: "2512.17547"
---

# G³Splat: Geometrically Consistent Generalizable Gaussian Splatting

<p align="center">
  <a href="https://m80hz.github.io/g3splat"><img src="https://img.shields.io/badge/🌐_Project_Page-007ACC?style=for-the-badge" alt="Project Page"></a>
  <a href="https://arxiv.org/abs/2512.17547"><img src="https://img.shields.io/badge/📄_Paper-B31B1B?style=for-the-badge" alt="arXiv"></a>
  <a href="https://github.com/m80hz/g3splat"><img src="https://img.shields.io/badge/💻_Code-181717?style=for-the-badge&logo=github" alt="GitHub"></a>
</p>

<p align="center">
  <img src="assets/teaser.png" alt="G³Splat Teaser" width="100%">
</p>

## Model Description

**G³Splat** is a pose-free, self-supervised framework for generalizable Gaussian splatting that achieves state-of-the-art performance in:
- 🎯 **Geometry Reconstruction** - Accurate depth and mesh reconstructions
- 📐 **Relative Pose Estimation** - No camera poses required at inference
- 🎨 **Novel View Synthesis** - High-quality image rendering from new viewpoints

## Available Checkpoints

| Model | Gaussian Type | Training Data | File |
|:------|:-------------:|:-------------:|:----:|
| G³Splat-3DGS | 3DGS | RealEstate10K | `g3splat_mast3r_3dgs_align_orient_re10k.ckpt` |
| G³Splat-2DGS | 2DGS | RealEstate10K | `g3splat_mast3r_2dgs_align_orient_re10k.ckpt` |

## Quick Start

```python
from huggingface_hub import hf_hub_download

# Download 3DGS model
ckpt_path = hf_hub_download(
    repo_id="g3splat/g3splat",
    filename="g3splat_mast3r_3dgs_align_orient_re10k.ckpt"
)

# Or download 2DGS model
ckpt_path_2dgs = hf_hub_download(
    repo_id="g3splat/g3splat",
    filename="g3splat_mast3r_2dgs_align_orient_re10k.ckpt"
)
```

## Usage

```bash
# Clone the repository
git clone https://github.com/m80hz/g3splat
cd g3splat

# Run demo
python demo.py --checkpoint pretrained_weights/g3splat_mast3r_3dgs_align_orient_re10k.ckpt
```
See the [GitHub repository](https://github.com/m80hz/g3splat) for full installation and usage instructions.

## Training Details

- **Training Data**: RealEstate10K
- **Resolution**: 256×256
- **Backbones**: MASt3R (ViT-Large) and VGGT
- **Hardware**: 24× A100 GPUs (6 nodes × 4 GPUs)
- **Training Time**: ~6 hours

## Citation

```bibtex
@inproceedings{g3splat,
  title     = {G3Splat: Geometrically Consistent Generalizable Gaussian Splatting},
  author    = {Hosseinzadeh, Mehdi and Chng, Shin-Fang and Xu, Yi and Lucey, Simon and Reid, Ian and Garg, Ravi},
  booktitle = {arXiv:2512.17547},
  year      = {2025},
  url       = {https://arxiv.org/abs/2512.17547}
}
```

## License

MIT License