CaltechAUTHORS
  A Caltech Library Service

Measuring and Predicting Object Importance

Spain, Merrielle and Perona, Pietro (2011) Measuring and Predicting Object Importance. International Journal of Computer Vision, 91 (1). pp. 59-76. ISSN 0920-5691. https://resolver.caltech.edu/CaltechAUTHORS:20110303-113534874

Full text is not posted in this repository. Consult Related URLs below.

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20110303-113534874

Abstract

How important is a particular object in a photograph of a complex scene? We propose a definition of importance and present two methods for measuring object importance from human observers. Using this ground truth, we fit a function for predicting the importance of each object directly from a segmented image; our function combines a large number of object-related and image-related features. We validate our importance predictions on 2,841 objects and find that the most important objects may be identified automatically. We find that object position and size are particularly informative, while a popular measure of saliency is not.


Item Type:Article
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.1007/s11263-010-0376-0DOIArticle
http://www.springerlink.com/content/m8nv11v601840402/PublisherArticle
http://rdcu.be/p3eBPublisherFree ReadCube access
http://resolver.caltech.edu/CaltechAUTHORS:CNS-TR-2007-002Related ItemTechnical Report
ORCID:
AuthorORCID
Perona, Pietro0000-0002-7583-5809
Additional Information:© 2011 Springer. Received: 9 December 2009; Accepted: 11 August 2010; Published online: 27 August 2010. This material is based upon work supported under a National Science Foundation Graduate Research Fellowship, Office of Naval Research grant N00014-06-1-0734, and National Institutes of Health grant R01 DA022777.
Funders:
Funding AgencyGrant Number
NSF Graduate Research FellowshipUNSPECIFIED
Office of Naval Research (ONR)N00014-06-1-0734
NIHR01 DA022777
Subject Keywords:Visual recognition; Object recognition; Importance; Perception; Keywording; Saliency; Rank aggregation; Amazon Mechanical Turk
Issue or Number:1
Record Number:CaltechAUTHORS:20110303-113534874
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20110303-113534874
Official Citation:Spain, M. & Perona, P. Int J Comput Vis (2011) 91: 59. doi:10.1007/s11263-010-0376-0
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:22637
Collection:CaltechAUTHORS
Deposited By: Jason Perez
Deposited On:04 Mar 2011 00:34
Last Modified:03 Oct 2019 02:39

Repository Staff Only: item control page