Architecture: Jean-Yves Bouguet/Intel PIRO: Benchmarking a Personal Image Retrieval System in ACES 2.402

Contact Name: 
Jenna Whitney
Date: 
Feb 6, 2006 11:00am - 12:00pm


There is a signup schedule for this event.

Speaker Name/Affiliation: Jean-Yves Bouguet/Intel

Talk Title: PIRO

: Benchmarking a Personal Image Retrieval System

Date/Time: Februar

y 6 2006 at 11:00 a.m.

Coffee: 10:30 a.m.

Location: ACES 2.

402

Host: Lizy John

Talk Abstract:
It is now common to hav

e accumulated tens of thousands of
personal pictures. Efficient access

to that many pictures
can only be done with a robust image retrieval sy

stem. This
application is of high interest to processor architects.
It is highly compute intensive and could motivate end users
to upgrad

e their personal computers to the next generations
of processors.
A key question is how to assess the robustness of a personal
image re

trieval system. Personal image databases are very
different from digita

l libraries that have been used by
many Content Based Image Retrieval S

ystems. Personal image
databases are very different from digital librar

ies that
have been used by many Content Based Image Retrieval Systems.

For example a personal image database has a lot of pictures
of peop

le but a small set of different people typically
family relatives an

d friends. Pictures are taken in a
limited set of places like home wor

k school and vacation
destination. The most frequent queries are sear

ched for
people and for places. These attributes and many others

affect how a personal image retrieval system should be benchmarked
and
benchmarks need to be different from existing ones based
on art images
or medical images for examples.

The attributes of the data set do

not change the list of
components needed for the benchmarking of such s

ystems as
specified in: data sets query tasks ground truth and
e

valuation measures.

This talk proposes a way to build these componen

ts to be
representative of personal image databases and of the corresp

onding
usage models.

Speaker Bio:
Jean-Yves Bouguet is a Sen

ior Researcher at Intel''s Microprocessor
Research Labs since 1999. He

received his diplome d''ingenieur
from the Ecole Superieure d''Ingenieu

rs en Electrotechnique
et Electronique (ESIEE Paris) in 1994 and the M

.S. and
Ph.D. degrees in Electrical Engineering from the California
Institute of Technology (Caltech) in 1994 and 1999 respectively.
Comp

uter Vision Computer Graphics are his main research
fields of interest

.

During his thesis work he has developed and patented a
simple
and inexpensive method for scanning objects using
shadows. Subsequentl

y he developed modeling techniques
that combine 3D geometry capture an

d scene reflectance acquisition
for realistic rendering of real and syn

thetic scenes with
complex shape and surface characteristics for which

he also
holds a patent. Jean-Yves has received a number of distinctive

awards including the J. Walker von Brimer award for extraordinary
a

ccomplishments in the field of 3D photography in 1999.
Recently his re

search focus has moved to applying computational
vision techniques to i

mage and video mining applications
with a special emphasis on search an

d retrieval in personal
image and video collections.