Gaussian Mixture Models Tutorial

 

I needed to implement an image segmentation algorithm, but did not have the requisite training data to reproduce state-of-the-art methods in the domain. This motivated me to study the application of Gaussian Mixture Models, a classical machine learning approach that – while unable to produce state-of-the-art results – is fully unsupervised and does not require training data. I wrote a mathematics and programming tutorial paper for computing Gaussian Mixture Models using the Expectation Maximization algorithm and applying the models to image segmentation. The PDF is available here, and the accompanying Python reference implementation is available on my GitHub page.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s