 |
    |
Spatio-temporal Event Classification Using Time-Series Kernel Based Structured Sparsity*
László A. Jeni1, András L
rincz2, Zoltán Szabó3, Jeffrey F. Cohn1, 4, and Takeo Kanade1
1Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, USA
laszlo.jeni@ieee.org
jeffcohn@cs.cmu.edu
tk@cs.cmu.edu
2Faculty of Informatics, Eötvös Loránd University, Budapest, Hungary
andras.lorincz@elte.hu
3Gatsby Computational Neuroscience Unit, University College London, London, UK
zoltan.szabo@gatsby.ucl.ac.uk
4Department of Psychology, University of Pittsburgh, Pittsburgh, PA, USA
Abstract. In many behavioral domains, such as facial expression and gesture, sparse structure is prevalent. This sparsity would be well suited for event detection but for one problem. Features typically are confounded by alignment error in space and time. As a consequence, high-dimensional representations such as SIFT and Gabor features have been favored despite their much greater computational cost and potential loss of information. We propose a Kernel Structured Sparsity (KSS) method that can handle both the temporal alignment problem and the structured sparse reconstruction within a common framework, and it can rely on simple features. We characterize spatio-temporal events as time-series of motion patterns and by utilizing time-series kernels we apply standard structured-sparse coding techniques to tackle this important problem. We evaluated the KSS method using both gesture and facial expression datasets that include spontaneous behavior and differ in degree of difficulty and type of ground truth coding. KSS outperformed both sparse and non-sparse methods that utilize complex image features and their temporal extensions. In the case of early facial event classification KSS had 10% higher accuracy as measured by F1 score over kernel SVM methods.
Keywords: structured sparsity, time-series kernels, facial expression classification, gesture recognition
Electronic Supplementary Material:
LNCS 8692, p. 135 ff.
lncs@springer.com
© Springer International Publishing Switzerland 2014