Speaker: Lior Wolf 
Title: Discriminative approaches for object recognition from few =
examples.
Friday 2/18/05 11am CoRE 301 

Abstract

In my talk I will first present a novel set of
features for robust object recognition, which
exhibits outstanding performance on a variety of
object categories while being capable of learning
from only few training examples. These features are
motivated by Poggio's quantitative model of
visual cortex. The system built on top of these
features outperforms state-of-the-art systems on a
variety of object image datasets.

I will then present and analyze a novel regularization
technique based on enhancing our dataset
with corrupted copies of the original data. The
motivation is that since the learning algorithm
lacks information about which parts of the data are
reliable, it has to produce more robust
classification functions. Using this framework, I will
propose a simple addition to the gentle
boosting algorithm which enables it to work with only
few training examples. I will conclude with
several experiments on vision and biological
datasets.

Bio: Lior Wolf is a post-doctoral associate in the
CBCL lab at MIT, managed by Prof. Tomaso Poggio.
He graduated from the Hebrew University in Israel,
where he worked under the supervision of Prof.
Amnon Shashua. His joint work with prof. Shashua
at ECCV 2000 received the best
paper award, and their work in ICCV 2001
received the Marr prize honorable mention. His
research interests include object recognition, video
analysis and machine learning algorithms.

-------------------------------------------------------