There is a sign up schedule for this event: http://www.cs.ut

exas.edu/department/webevent/utcs/events/cgi/eidshow.cgi?person=PedroF.Felz

enszwalb

Type of Talk:  UTCS Colloquium/AI

Speaker/Affili

ation:  Pedro F. Felzenszwalb/University of Chicago 

Date

/Time:  Friday, March 6, 2009  11:00 a.m.

Location:&nbs

p; ACES 2.402

Host:  Kristen Grauman

Title:  &quot

;Object Recognition with Part-Based Models"

Talk Abstract: &nb

sp;

Object recognition is one of the most important and challenging
problems in computer vision. One of the main difficulties is in developing
representations that can effectively capture typical variations in appeara

nce that occur within a class of objects. Deformable part-based models prov

ide a natural and elegant framework for addressing this problem. However,

they also lead to difficult computational problems. I will describe efficie

nt algorithms we have developed for finding objects in images using deforma

ble part-based models. I will also describe a set of techniques we have dev

eloped for training part-based models from weakly labeled data, using a fo

rmalism we call latent SVM. We have used these methods to implement a state

-of-the-art system for finding objects from a generic category, such as pe

ople or cars, in cluttered images.

Speaker Bio:

Pedro Felzens

zwalb is an associate professor in the Department of Computer Science at th

e University of Chicago. His research focuses on computer vision, algorith

ms and artificial intelligence. He received his PhD in computer science fro

m MIT in 2003.