A Caltech Library Service

Unsupervised Learning of Categorical Segments in Image Collections


Andreetto, Marco (2011) Unsupervised Learning of Categorical Segments in Image Collections. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/ZH04-VT55.


Which one comes first: segmentation or recognition? We propose a unified framework for carrying out the two simultaneously and without supervision. The framework combines a flexible probabilistic model for representing the shape and appearance of each segment, with the popular "bag of visual words" model for recognition. If applied to a collection of images, our framework can simultaneously discover the segments of each image, and the correspondence between such segments, without supervision. Such recurring segments may be thought of as the "parts" of corresponding objects that appear multiple times in the image collection. Thus, the model may be used for learning new categories, detecting/classifying objects, and segmenting images, without using expensive human annotation.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:Computer Vision, Machine Learning, Image Segmentation, Object Recognition, Statistical Models, Montecarlo Methods
Degree Grantor:California Institute of Technology
Division:Engineering and Applied Science
Major Option:Electrical Engineering
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Perona, Pietro
Thesis Committee:
  • Perona, Pietro (chair)
  • Abu-Mostafa, Yaser S.
  • Hassibi, Babak
  • Welling, Max
  • Belongie, Serge J.
Defense Date:12 January 2011
Record Number:CaltechTHESIS:04262011-213152111
Persistent URL:
Related URLs:
URLURL TypeDescription website
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:6355
Deposited By: Marco Andreetto
Deposited On:27 May 2011 20:34
Last Modified:08 Nov 2023 00:44

Thesis Files

PDF - Final Version
See Usage Policy.


Repository Staff Only: item control page