sofima

SOFIMA

SOFIMA (Scalable Optical Flow-based Image Montaging and Alignment) is a tool for stitching, aligning and warping large 2d, 3d and 4d microscopy datasets.

Please cite as https://zenodo.org/records/10534541

This is not an officially supported Google product.

Installation

SOFIMA is implemented purely in Python, and does not require a build step. To install it directly from the repository, run:

  pip install git+https://github.com/google-research/sofima

Overview

SOFIMA uses optical flow regularized with an elastic mesh to establish maps between data in different coordinate systems. Both the flow estimator as well as the mesh solver are implemented in JAX and will automatically take advantage of GPU acceleration if the hardware if available.

A core data structure used throughout the project is a coordinate map stored as a dense array of relative offsets (see the module docstring in map_utils.py for details). Among other uses, this is the representation of the estimated flow fields and the mesh node positions.

Example usage

• electron microscopy tile stitching
• electron microscopy section alignment
• LICONN 3d tile stitching

Citation

If you use this software in your research, please cite it using the following metadata:

Januszewski, M., Blakely, T., & Lueckmann, J.-M. (2024). SOFIMA: Scalable Optical Flow-based Image Montaging and Alignment. Zenodo. https://doi.org/10.5281/zenodo.10534541

License

Licensed under the Apache License, Version 2.0 (the “License”); you may not use this software except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0.

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an “AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.