logo Idiap Research Institute        
 [BibTeX] [Marc21]
Constructing visual models with a latent space approach
Type of publication: Idiap-RR
Citation: monay:rr05-14
Number: Idiap-RR-14-2005
Year: 2005
Institution: IDIAP
Note: To appear in Springer series of Lecture Notes in Computer Science
Abstract: We propose the use of latent space models applied to local invariant features for object classification. We investigate whether using latent space models enables to learn patterns of visual co-occurrence and if the learned visual models improve performance when less labeled data are available. We present and discuss results that support these hypotheses. Probabilistic Latent Semantic Analysis (PLSA) automatically identifies aspects from the data with semantic meaning, producing unsupervised soft clustering. The resulting compact representation retains sufficient discriminative information for accurate object classification, and improves the classification accuracy through the use of unlabeled data when less labeled training data are available. We perform experiments on a 7-class object database containing 1776 images.
Userfields: ipdmembership={vision},
Projects Idiap
Authors Monay, Florent
Quelhas, Pedro
Gatica-Perez, Daniel
Odobez, Jean-Marc
Crossref by monay:pascal:2006
Added by: [UNK]
Total mark: 0
  • monay-idiap-rr-05-14.pdf
  • monay-idiap-rr-05-14.ps.gz