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. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1109/ICCV.2011.6126445 | ICCV |
Keywords | Field | DocType |
object categorization,underlying grid structure,graph-matching kernel,nearby region,fast approximate algorithm,category-level image classification,geometric consistency,dense set,underlying image model,svm-based image classification,single type,scenes datasets,optimization,image retrieval,approximation algorithms,kernel,support vector machines,image classification,support vector machine,graph matching,vectors,graph theory,edge detection,geometry | Kernel (linear algebra),Graph theory,Approximation algorithm,Computer vision,Caltech 101,Pattern recognition,Computer science,Support vector machine,Image retrieval,Matching (graph theory),Artificial intelligence,Contextual image classification | Conference |
Volume | Issue | ISSN |
2011 | 1 | 1550-5499 |
Citations | PageRank | References |
131 | 3.20 | 32 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Olivier Duchenne | 1 | 382 | 10.34 |
Armand Joulin | 2 | 1683 | 61.97 |
Jean Ponce | 3 | 12182 | 902.31 |