MultiverSeg

Project Page | Paper

Official implementation of MultiverSeg: Scalable Interactive Segmentation of Biomedical Imaging Datasets with In-Context Guidance

Hallee E. Wong, Jose Javier Gonzalez Ortiz, John Guttag, Adrian V. Dalca

Updates

• (2025-01-26) inference code and weights released
• (2024-12-19) preprint released!

Models

We provide pre-trained weights here.

Installation

You can install multiverseg in two ways:

• With pip:

pip install git+https://github.com/halleewong/MultiverSeg.git

• Manually: cloning it and installing dependencies

git clone https://github.com/halleewong/MultiverSeg
python -m pip install -r ./MultiverSeg/requirements.txt
export PYTHONPATH="$PYTHONPATH:$(realpath ./MultiverSeg)"

Getting Started

First download the model checkpoints

cd checkpoints
./download.sh

Then see ./notebooks/inference.ipynb for a tutorial.

Acknowledgements

This project builds extensively on code originally developed for ScribblePrompt and UniverSeg

Citation

If you find our work or any of our materials useful, please cite our paper:

@article{wong2024multiverseg,
  title={MultiverSeg: Scalable Interactive Segmentation of Biomedical Imaging Datasets with In-Context Guidance},
  author={Hallee E. Wong and Jose Javier Gonzalez Ortiz and John Guttag and Adrian V. Dalca},
  journal={arXiv preprint arXiv:2412.15058},
  year={2024},
}