Scivision: A toolkit for scientific image analysis

The scivision project aims to connect computer vision model developers to image data providers from diverse scientific fields.

It’s core features are:

The scivision catalog, containing pretrained computer vision models and datasets from science and the humanities
The scivision model and datasource API, a simple, standard interface to models and data that works with the entries in the catalog

Start with the 👩‍💻 User guide for an introduction to using Scivision.

See an interactive demonstration of Scivision right now in your browser, no installation required:

Browse through a gallery of notebooks (runnable interactively with Binder) from research projects using Scivision at the Scivision Gallery.

Scivision development happens in the Scivision GitHub repo.

This Scivision documentation is hosted on Read the Docs.

📚 Contents

The Scivision project was conceived at The Alan Turing Institute, as an open software community that:

Empowers scientific domain experts to easily access and integrate the latest computer vision tools
Provides a bridge between computer vision algorithms and datasets from different academic disciplines
Enables algorithm developers and data providers to distribute their methods and datasets

Consider starting a GitHub discussion or opening an issue.

To contact the project team directly, email the Scivison maintainers.

The Scivision software is released under a BSD Licence by The Alan Turing Institute.