Graph cuts in computer vision: Difference between revisions

Revision as of 03:34, 11 August 2019

Revision as of 01:42, 2 November 2019

Line 39:

* Then, we use these histograms to set the regional penalties as negative log-likelihoods.

===== GMM (Gaussian Mixturemixture Modelmodel) =====

* We usually use 2two distributions: one for background modelling and another for foreground pixels.

* Use a Gaussian mixture model (with 5–8 components) to model those 2 distributions.

* Goal: Try to pull apart those 2two distributions.

===== Texon =====