---
license: cc-by-nc-4.0
base_model:
- stabilityai/stable-diffusion-xl-base-1.0
tags:
- image-to-image
inference: false
---
# ✨ Latent Bridge Matching for Normals Estimation ✨

Latent Bridge Matching (LBM) is a new, versatile and scalable method proposed in [LBM: Latent Bridge Matching for Fast Image-to-Image Translation](https://arxiv.org/abs/2503.07535) that relies on Bridge Matching in a latent space to achieve fast image-to-image translation.
This model was trained to estimate the normal map from a given input image.
See also our [live demo](https://huggingface.co/spaces/jasperai/LBM_relighting) for image relighting and official [Github repo](https://github.com/gojasper/LBM).

## How to use?
To use this model you need first to install the associated `lbm` library by running the following
```bash
pip install git+https://github.com/gojasper/LBM.git
```

Then, you can infer with the model on your input images 
```python
import torch
from PIL import Image
from lbm.inference import evaluate, get_model

# Load model
model = get_model(
  "jasperai/LBM_relighting",
  torch_dtype=torch.bfloat16,
  device="cuda",
)

# Load a source image
source_image = Image.open("your_image")

# Perform inference
output_image = evaluate(model, source_image, num_sampling_steps=1)
```

## Metrics
This model achieves the following metrics

| Metrics | mean ↓ | 11.25 ↑ | 30.0 ↑ |
|---------|--------|---------|--------|
| NYUv2   | 15.5   | 62.5    | 84.9   |
| ScanNet | 14.1   | 65.8    | 87.0   |
| iBims   | 16.9   | 68.3    | 82.7   |
| Sintel  | 32.2   | 24.0    | 58.6   |

## License
This code is released under the Creative Commons BY-NC 4.0 license.

## Citation
If you find this work useful or use it in your research, please consider citing us
```bibtex
@article{chadebec2025lbm,
      title={LBM: Latent Bridge Matching for Fast Image-to-Image Translation}, 
      author={Clément Chadebec and Onur Tasar and Sanjeev Sreetharan and Benjamin Aubin},
      year={2025},
      journal = {arXiv preprint arXiv:2503.07535},
}
```