A group sparsity-driven approach to 3-D action recognition

Coşar, Serhan and Çetin, Müjdat (2011) A group sparsity-driven approach to 3-D action recognition. In: IEEE International Conference on Computer Vision, Workshop on Visual Surveillance, Barcelona, Spain

[thumbnail of cosar_ICCV_VS11.pdf] PDF
cosar_ICCV_VS11.pdf

Download (1MB)

Abstract

In this paper, a novel 3-D action recognition method based on sparse representation is presented. Silhouette images from multiple cameras are combined to obtain motion history volumes (MHVs). Cylindrical Fourier transform of MHVs is used as action descriptors. We assume that a test sample has a sparse representation in the space of training samples. We cast the action classification problem as an optimization problem and classify actions using group sparsity based on l1 regularization. We show experimental results using the IXMAS multi-view database and demonstratethe superiority of our method, especially when observations are low resolution, occluded, and noisy and when the feature dimension is reduced.
Item Type: Papers in Conference Proceedings
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Electronics
Faculty of Engineering and Natural Sciences
Depositing User: Müjdat Çetin
Date Deposited: 06 Jan 2012 10:54
Last Modified: 26 Apr 2022 09:04
URI: https://research.sabanciuniv.edu/id/eprint/18332

Actions (login required)

View Item
View Item