Seminar Details

Seminar Details


Mar 3

2:30 pm

Learning the \"Epitome\" of an Image

Brendan J. Frey


University of Toronto

I will describe a new model of image data that we call the "epitome".
The epitome of an image is its miniature, condensed version containing the essence of the textural and shape properties of the image. As opposed to previously used simple image models, such as templates or basis functions, the size of the epitome is considerably smaller than the size of the image or object it represents, but the epitome still contains most constitutive elements needed to reconstruct the image. A collection of images often shares an epitome, e.g., when images are a few consecutive frames from a video sequence, or when they are photographs of similar objects. A particular image in a collection is defined by its epitome and a smooth mapping from the epitome to the image pixels. When the epitome model is used within a hierarchical generative model, appropriate inference algorithms can be derived to extract epitomes from a single image or a collection of images and at the same time perform various inference tasks, such as image segmentation, motion estimation, object removal, super-resolution and image denoising.

Go to for a sneak preview.

Joint work with Nebojsa Jojic and Anitha Kannan.


Brendan J. Frey was born on August 29, 1968, in Calgary, Alberta near the foothills of the Rocky Mountains, where he enjoyed hiking and camping with his family. In 1979, he started writing computer programs, attaching sensors to his home computer, and building simple robots. His first publication (Run Magazine, 1981) describes software for simulating a generative model of images, where the pixel intensities are independent. His academic education was in the areas of physics, engineering and computer science, culminating with a doctorate from Geoffrey Hinton\'s Neural Networks Research Group at the University of Toronto in 1997. From 1997 to 1999, Frey was a Beckman Fellow at the University of Illinois at Urbana-Champaign, where he continues to be an adjunct faculty member in Electrical and Computer Engineering. From 1998 to 2001, he was a faculty member in Computer Science at the University of Waterloo. Currently, Frey is head of the Probabilistic and Statistical Inference Group, in the Department of Electrical and Computer Engineering at the University of Toronto. Frey consults for Microsoft Research Redmond, and various start-up companies in the Toronto-Waterloo area.

He has received several awards, including the Premier\'s Research Excellence Award of Canada, and he has given over 40 invited talks and published over 100 papers on inference and estimation in complex probability models, with applications in machine learning, computer vision, molecular biology, audio processing, image processing and iterative decoding. In 2003, he co-chaired the Canadian Workshop on Information Theory and the Workshop on Artificial Intelligence and Statistics. He was a Co-Editor-in-Chief of the February 2000 special issue of the IEEE Transactions on Information Theory, titled Codes on Graphs and Iterative Algorithms, and is currently an Associate Editor of the IEEE Transactions on Pattern Analysis and Machine Intelligence.