Skip to content (access key 's')
Logo of Technion
Logo of CS Department


Pixel Club: Natural Images, Gaussian Mixtures and Dead Leaves
event speaker icon
Daniel Zoran (Hebrew University of Jerusalem)
event date icon
Tuesday, 18.6.2013, 11:30
event location icon
Room 337-8 Taub Bld.
Simple Gaussian Mixture Models (GMMs) learned from pixels of natural image patches have been recently shown to be surprisingly strong performers in modeling the statistics of natural images. Here we provide an in depth analysis of this simple yet rich model. We show that such a GMM model is able to compete with even the most successful models of natural images in log likelihood scores, denoising performance and sample quality. We provide an analysis of what such a model learns from natural images as a function of number of mixture components --- including covariance structure, contrast variation and intricate structures such as textures, boundaries and more. Finally, we show that the salient properties of the GMM learned from natural images can be derived from a simplified Dead Leaves model which explicitly models occlusion, explaining its surprising success relative to other models.
[Back to the index of events]