The Menpo project is a set of Python libraries for manipulating data
that is particularly useful for Machine Learning and Computer Vision. The core
menpo library contains a number of powerful tools for manipulating
annotated image and mesh data.
menpo focuses on making importing,
manipulating and visualizing data as simple as possible. This makes us strong
supporters of projects like the Jupyter notebook.
The Menpo project also includes another tool that makes annotating new datasets even easier! Freely available a landmarker.io, the landmarker is a tool that makes it simple to annotate new images.
On top of the core
menpo library, we have implemented a number of other
libraries that utilize
menpo for more specific purposes. Currently, the
Menpo project libraries include:
menpo: the core library containing all the essential classes for data manipulation including transforms, warping, landmarks, importing and exporting.
menpofit: a statistical modelling toolkit, providing all the tools required to build, fit and test deformable models like Active Appearance Models, Constrained Local Models and Supervised Descent Method.
menpo3d: specific tools for visualizing and manipulating 3D data such as meshes. This package contains a lot of complex dependencies, including an OpenGL based rasterizer.
menpodetect: wraps a number of existing projects that can perform object detection. Not all of the wrapped projects fall under the same BSD license and so care must be taken when using this project to adhere to the sub-project licenses.
menpowidgets: provides Jupyter notebook interactive widgets. These widgets are for all of the Menpo project subprojects including
The Menpo project is written in Python and we provide a simple and easy
method of installation using Conda. We suggest you
head over to the
installation instructions to get
started. We recommend the use of
conda due to the fact that the Menpo projects also include compiled
C/C++ code which may be complicated to compile on various platforms.
Joan Alabort-i-Medina, Epameinondas Antonakos, James Booth, Patrick Snape and Stefanos Zafeiriou, "Menpo: A comprehensive platform for parametric image alignment and visual deformable models", In Proceedings of the ACM International Conference on Multimedia, MM ’14, pages 679-682, New York, NY, USA, 2014. ACM.[Bibtex] [pdf]