There is a sign-up schedule for this event that can be found at
http://www.cs.utexas.edu/department/webevent/utcs/events/cgi/list_events.c

gi

Type of Talk: UTCS Colloquium/AI

Speaker/Affiliation: Ian En

dres/UIUC

Date/Time: Friday, August 6, 2010 12:30 p.m.

Locati

on: ACES 3.408 Host: Kristen Grauman

Talk Title: Category Independent
Object Proposals

Talk Abstract: We propose a category-independent me

thod to produce a bag of regions and rank them, such that top-ranked regio

ns are likely to be good segmentations of different objects. Our key objec

tives are completeness and diversity: every object should have at least one
good proposed region, and a diverse set should be top-ranked. Our approac

h is to generate a set of segmentations by performing graph cuts based on a
seed region and a learned affinity function. Then, the regions are ranked
using structured learning based on various cues. Our experiments on BSDS a

nd PASCAL VOC 2008 demonstrate our ability to