CaltechAUTHORS
  A Caltech Library Service

Entropy-based active learning for object recognition

Holub, Alex and Perona, Pietro and Burl, Michael C. (2008) Entropy-based active learning for object recognition. In: Computer Vision and Pattern Recognition Workshops, 2008. CVPRW '08. IEEE Computer Society Conference on : 23-28 June 2008. No.Procee. IEEE , Piscataway, NJ, pp. 885-892. ISBN 978-1-4244-2339-2. https://resolver.caltech.edu/CaltechAUTHORS:20100629-110925891

[img]
Preview
PDF - Published Version
See Usage Policy.

557kB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20100629-110925891

Abstract

Most methods for learning object categories require large amounts of labeled training data. However, obtaining such data can be a difficult and time-consuming endeavor. We have developed a novel, entropy-based ldquoactive learningrdquo approach which makes significant progress towards this problem. The main idea is to sequentially acquire labeled data by presenting an oracle (the user) with unlabeled images that will be particularly informative when labeled. Active learning adaptively prioritizes the order in which the training examples are acquired, which, as shown by our experiments, can significantly reduce the overall number of training examples required to reach near-optimal performance. At first glance this may seem counter-intuitive: how can the algorithm know whether a group of unlabeled images will be informative, when, by definition, there is no label directly associated with any of the images? Our approach is based on choosing an image to label that maximizes the expected amount of information we gain about the set of unlabeled images. The technique is demonstrated in several contexts, including improving the efficiency of Web image-search queries and open-world visual learning by an autonomous agent. Experiments on a large set of 140 visual object categories taken directly from text-based Web image searches show that our technique can provide large improvements (up to 10 x reduction in the number of training examples needed) over baseline techniques.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.1109/CVPRW.2008.4563068 DOIUNSPECIFIED
ORCID:
AuthorORCID
Perona, Pietro0000-0002-7583-5809
Additional Information:U.S. Government work not protected by U.S. copyright. This research has been carried out in part at the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration (NASA). Funding provided under the Applied Information Systems Research Program. A.H. and P.P. were supported by N00014-06-1- 0734.
Funders:
Funding AgencyGrant Number
Applied Information Systems Research ProgramN00014-06-1-0734
Subject Keywords:Web image-search queries , autonomous agent , entropy-based active learning , object categories , object recognition , open-world visual learning , text-based Web image , unlabeled images
Other Numbering System:
Other Numbering System NameOther Numbering System ID
INSPEC Accession Number10104399
Issue or Number:Procee
DOI:10.1109/CVPRW.2008.4563068
Record Number:CaltechAUTHORS:20100629-110925891
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20100629-110925891
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:18852
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:14 Jul 2010 16:56
Last Modified:08 Nov 2021 23:47

Repository Staff Only: item control page