Segment Anything Based Electron tomography Recognition is a robust platform designed for autonomous segmentation of organelles from cryo-electron tomography (cryo-ET) or electron microscopy (EM) datasets.
Leveraging foundational models, SABER enables segmentation directly from video-based training translated into effective 3D tomogram analysis. Users can utilize zero-shot inference with morphological heuristics or enhance prediction accuracy through data-driven training.
- 🔍 Zero-shot segmentation: Segment EM/cryo-ET data without explicit retraining, using foundational vision models.
- 🖼️ Interactive GUI for labeling: Intuitive graphical interface for manual annotation and segmentation refinement.
- 🧠 Expert-driven classifier training: Fine-tune segmentation results by training custom classifiers on curated annotations.
- 🧊 3D organelle segmentation: Generate volumetric segmentation masks across tomographic slices.
Saber is available on PyPI and can be installed using pip:
pip install saber-em- By default, the GUI is not included in the base installation. To enable the graphical interface for manual annotation, install with:
pip install saber-em[gui]- One of the current dependencies is currently not working with pip 25.1. We recommend using pip 25.2 or higher when installing saber:
pip install --upgrade "pip>=25.2"SABER provides a clean, scriptable command-line interface. Run the following command to view all available subcommands:
saber --help
For detailed documentation, tutorials, CLI and API reference, visit our documentation
This project adheres to the Contributor Covenant code of conduct. By participating, you are expected to uphold this code. Please report unacceptable behavior to [email protected].
If you believe you have found a security issue, please responsibly disclose by contacting us at [email protected].