SAMJ: Fast Image Annotation on ImageJ/Fiji via Segment Anything Model

Published in Nature Communications, 2026

Recommended citation: García-López-de-Haro, C., Fuster-Barceló, C., Rueden, C. T., Heras, J., Ulman, V., Franco-Barranco, D., Inés, A., Eliceiri, K. W., Olivo-Marin, J.-C., Tinevez, J.-Y., Sage, D., & Muñoz-Barrutia, A. (2026). SAMJ: fast image annotation on ImageJ/Fiji via segment anything model. Nature Communications, 17, 5402. https://doi.org/10.1038/s41467-026-71752-x https://www.nature.com/articles/s41467-026-71752-x

SAMJ: Fast Image Annotation on ImageJ/Fiji via Segment Anything Model

Published in: Nature Communications
Authors: Carlos García-López-de-Haro, Caterina Fuster-Barceló, Curtis T. Rueden, Jónathan Heras, Vladimír Ulman, Daniel Franco-Barranco, Adrián Inés, Kevin W. Eliceiri, Jean-Christophe Olivo-Marin, Jean-Yves Tinevez, Daniel Sage, Arrate Muñoz-Barrutia

Abstract

SAMJ is a plugin that integrates the Segment Anything Model (SAM) into the Fiji/ImageJ ecosystem, empowering biologists with a fast and intuitive tool for high-quality image annotation.

Built to reduce technical barriers, SAMJ uses Java-Python integration (via Appose) for seamless installation and real-time segmentation with various SAM variants—such as SAM-2, EfficientSAM, and EfficientViTSAM—even on modest hardware.

Key Features

  • Interactive prompt-based segmentation
  • One-click installer for ImageJ/Fiji users
  • Live and batch annotation modes
  • Macro support and integration with existing Fiji tools
  • Compatibility with BigDataViewer and Labkit for 3D or multi-label tasks
  • Application areas: Nuclei segmentation, tumor quantification, electron microscopy, and more

This plugin significantly lowers the entry barrier to modern segmentation techniques, enabling accessible and reproducible AI-assisted annotation pipelines for life sciences and medical imaging.

👉 Read the published article in Nature Communications