This is the official homepage of the Gamera project, a framework for building document analysis applications.

About Gamera

Gamera is not a packaged document recognition system, but a toolkit for building document image recognition systems. It makes the development of new recognition system quite easy, though this still requires some time commitment. Gamera is a cross platform library for the Python programming language. Apart from providing a set of commonly needed functionality for document image analysis, Gamera additionally allows for custom extensions as C++ or Python Plugins and as Toolkits.

For an overview both of Gamera itself and of applications built with the aid of Gamera, see the papers listed under "Publications". Ready to run recognition systems for custom application cases and other optional add-on libraries are listed under "Addons". The addons are distributed as toolkits.


The official Gamera documentation describes most aspects of Gamera like installation, the training GUI, image types and plugins and writing custom scripts and toolkits. It comes with the source distribution of Gamera (see the doc subdirectory). Here are direct links to useful documentation:

If you are new to the Python language, you might also want to check some of the documentation from the Python homepage.


Please consult the Gamera FAQ first to see whether your question is an already well known problem. For bug reports, you can use the respective github project for posting a bug or for suggesting a fix as a pull request. For general question, see "Authors" below for possible persons to ask.

Help is always welcome, whether it's improving the documentation, bug reports, coding new features or offering design suggestions. There is a TODO list. This may provide inspiration for people looking for projects to do.

Authors and Acknowledgements

The Gamera project was started 2001 by Michael Droettboom, Karl Mac Millan and Ichiro Fujinaga at the "Digital Knowledge Center" of the Johns Hopkins University (USA). In 2006, Christoph Dalitz from the Niederrhein University of Applied Sciences (Germany) joined the development team and is responsible since 2008, together with Michael Droettboom, for the maintenance of Gamera. For more information, see Michael Droettboom's brief history of the Gamera document image analysis system.

Today, the primary developers of Gamera are:

Former primary developers (these people no longer work on Gamera as part of their "day job"):

Thanks to the following people who have contributed their time, support, feedback or code to Gamera (in alphabetical order):

Teal Anderson, Grant Arnold, Colin Baumgarten, Tobias Bolten, Christian Brandt, Robert Butz, Lukas Cenovsky, Oliver Christen, Alexander Cobb, Jens Dahlmanns, Georg Drees, Tim DiLauro, Robert Ferguson, Andrew Hankinson, Manuel Jeltsch, Thomas Karsten, Jürgen Killian, David Kolanus, Uma Kompella, Jonathan Koch, Jean-Francois Leon, Andreas Leuschner, David Lewis, Moritz Muras, Mark Patton, Raghu, Jason Riesa, Stefan Ruloff, Seo Sanghyeon, Vincent Saulnier, Markus Schaaf, Fabian Schmitt, Anton Shkolnik, Daniel Stender, Marcel Schroers, Julian Tillmanns, Daveed Vandevoorde, Jens Wilberg, Jakub Wilk, Hasan Yildiz, Jochen Zimmerman

Parts of the project were funded by the US National Science Foundation, the Institute for Museum and Library Services, and the Levy family.


Gamera is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 2 of the License, or (at your option) any later version.

Gamera is distributed in the hope that it will be useful, but without any warranty; without even the implied warranty of merchantability or fitness for a particular purpose. See the GNU General Public License for more details.