[Lisa_seminaires] Fwd: Séminaire demain, 3 juillet - École Poly, M-2110 @ 14h

Dumitru Erhan dumitru.erhan at umontreal.ca
Ven 3 Juil 11:28:53 EDT 2009


---------- Forwarded message ----------
From: Chris Pal <christopher.pal at polymtl.ca>
Date: 2009/7/2
Subject: Séminaire demain, 3 juillet - École Poly, M-2110 @ 14h
To: Dumitru Erhan <dumitru.erhan at umontreal.ca>


Bonjour,

Le 3 juillet à École Polytechnique, M-2110 @ 14h, Nicholas Morsillo étudiant
au doctorat à l'université de Rochester qui travaille avec moi présentera
son article dans IJCAI 2009 intitulé:
"Semi-Supervised Learning of Visual Classifiers from Web Images and Text"

Vous êtes cordialement invité à assister à la présentation et à échanger
avec Nick suite à la présentation.  La présentation elle-même sera d'environ
25 minutes.

- Chris Pal

Title of Talk:
Semi-Supervised Learning of Visual Classifiers from Web Images and Text
By: Nicholas Morsillo
To appear in: Proceedings of the Twenty-First International Conference on
Artificial Intelligence

Abstract: The web holds tremendous potential as a source of training data
for visual classification.  However, web images must be correctly indexed
and labeled before this potential can be realized.  Accordingly, there has
been considerable recent interest in collecting imagery from the web using
image search engines to build databases for object and scene recognition
research.  While search engines can provide rough sets of image data,
results are noisy and this leads to problems when training classifiers.  In
this paper we propose a semi-supervised model for automatically collecting
clean example imagery from the web.  Our approach includes both visual and
textual web data in a unified framework.  Minimal supervision is enabled by
the selective use of generative and discriminative elements in a
probabilistic model and a novel learning algorithm.  We show through
experiments that our model discovers good training images from the web with
minimal manual work.  Classifiers trained using our method significantly
outperform analogous baseline approaches on the Caltech-256 dataset.


----

Christopher J. Pal, Ph. D.
Professeur adjoint
Département de génie informatique et génie logiciel
Ecole Polytechnique de Montréal
Téléphone : (514) 340-5121, poste 7174


Plus d'informations sur la liste de diffusion Lisa_seminaires