On the discriminative power of learned vs. hand-crafted features for crowd density analysis

Marnissi, Mohamed Amine; Fradi, Hajer; Dugelay, Jean-Luc
IJCNN 2019, International Joint Conference on Neural Networks, July 14-19, 2019, Budapest, Hungary

Crowd density analysis is a crucial component in video surveillance mainly for security monitoring. This paper proposes a novel approach for crowd density classification, in
which learned features substitute the commonly used handcrafted features. In particular, the approach consists of employing deep networks to extract useful crowd features that can further be manageable by a classifier. This process is favorable for crowd features extraction due to the large learning capability of deep networks compared to traditional methods based on handcrafted features. The proposed approach is evaluated on three
challenging datasets, and the results demonstrate the effectiveness of learned features for crowd density classification. Furthermore, we include an extensive comparative study between different learned/hand-crafted features in order to investigate their discriminative
power to handle such problems. Their performance is evaluated using different classifiers and strategies as well.

DOI
Type:
Conference
City:
Budapest
Date:
2019-07-14
Department:
Digital Security
Eurecom Ref:
5953
Copyright:
© 2019 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

PERMALINK : https://www.eurecom.fr/publication/5953