A factorized model for multiple SVM and multi-label classification for large scale multimedia indexing
Résumé
This paper presents a set of improvements for SVM-based large scale multimedia indexing. The proposed method is particularly suited for the detection of many target concepts at once and for highly imbalanced classes (very infrequent concepts). The method is based on the use of multiple SVMs (MSVM) for dealing with the class imbalance and on some adaptations of this approach in order to allow for an efficient implementation using optimized linear algebra routines. The implementation also involves hashed structures allowing the factorization of computations between the multiple SVMs and the multiple target concepts, and is denoted as Factorized-MSVM.
Experiments were conducted on a large-scale dataset, namely TRECVid 2012 semantic indexing task. Results show that the Factorized-MSVM performs as well as the original MSVM, but it is significantly much faster. Speed-ups by factors of several hundreds were obtained for the simultaneous classification of 346 concepts, when compared to the original MSVM implementation using the popular libSVM implementation.
Mots clés
Training
Multimedia indexing and retrieval
Predictive models
Multimedia communication
Kernel
Euclidean distance
optimized linear algebra routines
multiple target concepts
multiple SVM
multimedia retrieval
multilabel classification
large-scale dataset
hashed structures
factorized-MSVM
factorized model
computations factorization
class imbalance
TRECVid 2012 semantic indexing task
SVM-based large scale multimedia indexing
video retrieval
support vector machines
multimedia databases
linear algebra
image classification
indexing
multi-learner approach
support vector machine
Domaines
Recherche d'information [cs.IR]| Origine | Fichiers produits par l'(les) auteur(s) |
|---|
Loading...