A Caltech Library Service

Unsupervised Organization of Image Collections: Taxonomies and Beyond

Bart, Evgeniy and Welling, Max and Perona, Pietro (2011) Unsupervised Organization of Image Collections: Taxonomies and Beyond. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33 (11). pp. 2302-2315. ISSN 0162-8828.

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

Use this Persistent URL to link to this item:


We introduce a nonparametric Bayesian model, called TAX, which can organize image collections into a tree-shaped taxonomy without supervision. The model is inspired by the Nested Chinese Restaurant Process (NCRP) and associates each image with a path through the taxonomy. Similar images share initial segments of their paths and thus share some aspects of their representation. Each internal node in the taxonomy represents information that is common to multiple images. We explore the properties of the taxonomy through experiments on a large (~ 10^4) image collection with a number of users trying to locate quickly a given image. We find that the main benefits are easier navigation through image collections and reduced description length. A natural question is whether a taxonomy is the optimal form of organization for natural images. Our experiments indicate that although taxonomies can organize images in a useful manner, more elaborate structures may be even better suited for this task.

Item Type:Article
Related URLs:
URLURL TypeDescription
Perona, Pietro0000-0002-7583-5809
Additional Information:© 2011 IEEE. Manuscript received 6 Apr. 2009; revised 29 Jan. 2010; accepted 2 Nov. 2010; published online 18 Apr. 2011. Recommended for acceptance by A. Torralba. This material is based upon work supported by the US National Science Foundation under Grant Nos. 0447903, 0535278 and IIS-0535292, and by US Office of Naval Research MURI grant 00014-06-1-0734. An early version of this paper appeared in [18]. The authors would like to thank Marco Andreetto for useful suggestions.
Funding AgencyGrant Number
Office of Naval Research (ONR)00014-06-1-0734
Subject Keywords:Taxonomy; hierarchy; clustering
Issue or Number:11
Record Number:CaltechAUTHORS:20110930-153331547
Persistent URL:
Official Citation:Bart, Evgeniy; Welling, Max; Perona, Pietro; , "Unsupervised Organization of Image Collections: Taxonomies and Beyond," Pattern Analysis and Machine Intelligence, IEEE Transactions on , vol.33, no.11, pp.2302-2315, Nov. 2011 doi: 10.1109/TPAMI.2011.79 URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:25517
Deposited By: Tony Diaz
Deposited On:06 Oct 2011 16:09
Last Modified:03 Oct 2019 03:06

Repository Staff Only: item control page