LBM Normals

✨ Latent Bridge Matching for Normals Estimation ✨

Latent Bridge Matching (LBM) is a novel, versatile, and scalable approach presented in LBM: Latent Bridge Matching for Fast Image-to-Image Translation. It leverages Bridge Matching in the latent space to enable rapid image-to-image translation. This model is trained to estimate the normal map from a given input image. Check out our live demo for image relighting and the official Github repo.

🚀 Quick Start

To use this model, you first need to install the associated lbm library by running the following command:

pip install git+https://github.com/gojasper/LBM.git

Then, you can perform inference on your input images using the model:

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)

💻 Usage Examples

Basic Usage

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)

📚 Documentation

Metrics

This model achieves the following metrics:

📄 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:

@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},
}