Paper Info
Title
Multivariate log-Gaussian Cox models of elementary shapes for recognizing natural scene categories
Abstract
In this paper, we address invariant scene classification from images. We propose a novel descriptor based on the statistical characterization of the spatial patterns formed by elementary objects in images. Elementary objects are defined from a tree of shapes of the topology map of the image and each object is characterized by shape context feature vector. Viewing the set of elementary objects as a realization of a random spatial process, we investigate a statistical analysis using log- Gaussian Cox model to define an invariant image descriptor. An application to natural scene recognition is described. Re- ported results validate the proposed descriptor with respect to previous work.
Year
DOI
Venue
2011
10.1109/ICIP.2011.6116640
ICIP
Keywords
Field
DocType
inner-distance shape context,random processes,trees (mathematics),shape context feature vector,statistical analysis,natural scene category recognition,image recognition,spatial pattern characterization,statistical characterization,image classification,elementary shapes,topographic map,gaussian processes,natural scenes,log-gaussian cox process,scene recognition,random spatial process realization,solid modelling,elementary objects,invariant scene classification,multivariate log-gaussian cox model,topology map,invariant image descriptor,cox model,visualization,cox process,probabilistic logic,feature vector,shape,correlation,spatial pattern
Computer vision,Feature vector,Pattern recognition,Computer science,Visualization,Gaussian,Invariant (mathematics),Artificial intelligence,Gaussian process,Probabilistic logic,Contextual image classification,Shape context
Conference
ISSN
ISBN
Citations 
1522-4880 E-ISBN : 978-1-4577-1302-6
978-1-4577-1302-6
1
PageRank 
References 
Authors
0.35
4
3
Name
Order
Citations
PageRank
Huu-Giao Nguyen1213.14
Ronan Fablet231247.04
Jean-Marc Boucher313222.28