ENGLISH
La vitrine de diffusion des publications et contributions des chercheurs de l'ÉTS
RECHERCHER

Adaptive skew-sensitive fusion of ensembles and their application to face re-identification

Téléchargements

Téléchargements par mois depuis la dernière année

Plus de statistiques...

De La Torre, Miguel, Granger, Éric et Sabourin, Robert. 2015. « Adaptive skew-sensitive fusion of ensembles and their application to face re-identification ». In International Joint Conference on Neural Networks 2015 (Killarney, Ireland, July 12-17, 2015)

[thumbnail of Granger E. 2015 11313 Adaptive skew-sensitive fusion of ensembles and their application.pdf]
Prévisualisation
PDF
Granger E. 2015 11313 Adaptive skew-sensitive fusion of ensembles and their application.pdf

Télécharger (1MB) | Prévisualisation

Résumé

Adaptive classifier ensembles have been shown to improve the accuracy and robustness of systems for face recognition (FR) in video surveillance. However, it is often assumed that the proportions of faces captured for target and non-target individuals are balanced, or they are known a priori, and constant over time. Some active approaches have been proposed to update the ensemble during operations according to class imbalance of the input data stream. Beyond the estimation operational class imbalance, these approaches commonly generate diverse pools of classifiers by selecting balanced training data, limiting the potential diversity provided by the abundant non-target data. In this paper, a skew-sensitive ensemble is proposed to adaptively combine classifiers trained with data selected to have varying levels of imbalance and complexity. Given a face re-identification application, faces captured for each person appearing in the scene are tracked and regrouped into trajectories. During enrollment, faces in a reference trajectory are combined with those of selected non-target trajectories to generate a pool of 2-class classifiers using data with various levels of imbalance and complexity. During operations, the level of imbalance is periodically estimated by comparing input trajectories and pre-computed histograms using Hellinger distance quantification. Ensemble fusion functions are then adapted based on the imbalance and complexity of operational data. Finally, ensemble scores are accumulated over trajectories for robust spatio-temporal FR. Results obtained in experiments with synthetic data and Face in Action videos reveal that the proposed approach can significantly improve performance across operational imbalances.

Type de document: Compte rendu de conférence
Professeur:
Professeur
Granger, Éric
Sabourin, Robert
Affiliation: Génie de la production automatisée, Génie de la production automatisée
Date de dépôt: 21 sept. 2015 17:47
Dernière modification: 21 août 2018 18:42
URI: https://espace2.etsmtl.ca/id/eprint/11313

Actions (Authentification requise)

Dernière vérification avant le dépôt Dernière vérification avant le dépôt