Skip to main content

A minimal ctypes-based port of some vlfeat functions.

Project description

This is a minimal port of a few components of the matlab interface of the vlfeat library for computer vision to Python.

vlfeat’s core library is written in C. In practice, though, a significant amount of the library lies in the MATLAB interface. This project is a port of a few functions from that interface to python/numpy, using ctypes. It contains only a few functions (the ones needed for py-sdm). The process isn’t very hard, it just takes some effort. Patches adding additional functionality are welcome.

There’s also a fork of vlfeat floating around that includes Boost::Python wrappers. I couldn’t get it to work, and didn’t try too hard because I saw that some of the functions I needed had significant amounts of matlab code anyway. You might be more interested in it, though; Andreas Mueller’s version appears to be the most recently updated.


The package can be installed by pip or easy_install normally. However, in order to actually use it, you’ll also need to download the vlfeat binary library. You can either install (or your platform’s equivalent) somewhere where ctypes can find it yourself, or use the included script for doing so; run it with python -m If you add a -h argument, it’ll show you how to do it with a pre-downloaded binary distribution. If you install Python packages with sudo, you may need to do the same for the download script.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

vlfeat-ctypes-0.1.5.tar.gz (30.8 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page