Title :
A graph-matching kernel for object categorization
Author :
Duchenne, Olivier ; Joulin, Armand ; Ponce, Jean
Author_Institution :
INRIA, Sophia Antipolis, France
Abstract :
This paper addresses the problem of category-level image classification. The underlying image model is a graph whose nodes correspond to a dense set of regions, and edges reflect the underlying grid structure of the image and act as springs to guarantee the geometric consistency of nearby regions during matching. A fast approximate algorithm for matching the graphs associated with two images is presented. This algorithm is used to construct a kernel appropriate for SVM-based image classification, and experiments with the Caltech 101, Caltech 256, and Scenes datasets demonstrate performance that matches or exceeds the state of the art for methods using a single type of features.
Keywords :
geometry; graph theory; image classification; image matching; support vector machines; SVM-based image classification; category-level image classification; fast approximate algorithm; geometric consistency; graph-matching kernel; grid structure; image model; object categorization; Approximation algorithms; Image edge detection; Image retrieval; Kernel; Optimization; Support vector machines; Vectors;
Conference_Titel :
Computer Vision (ICCV), 2011 IEEE International Conference on
Conference_Location :
Barcelona
Print_ISBN :
978-1-4577-1101-5
DOI :
10.1109/ICCV.2011.6126445