This paper proposes a new approach to model the temporal dynamics of a sequence of facial expressions. To this purpose, a sequence of Face Image Descriptors (FID) is regarded as the output of a Linear Time Invariant (LTI) system. The temporal dynamics of such sequence of descriptors are represented by means of a Hankel matrix. The paper presents different strategies to compute dynamics-based representation of a sequence of FID, and reports classification accuracy values of the proposed representations within different standard classification frameworks. The representations have been validated in two very challenging application domains: emotion recognition and pain detection. Experiments on two publicly available benchmarks and comparison with state-of-the-art approaches demonstrate that the dynamics-based FID representation attains competitive performance when off-the- shelf classification tools are adopted.
Lo Presti, L., La Cascia, M. (2015). Using Hankel matrices for dynamics-based facial emotion recognition and pain detection. In Computer Vision and Pattern Recognition Workshops (CVPRW), 2015 IEEE Conference on (pp. 26-33). IEEE [10.1109/CVPRW.2015.7301351].
Using Hankel matrices for dynamics-based facial emotion recognition and pain detection
LO PRESTI, Liliana
;LA CASCIA, Marco
2015-01-01
Abstract
This paper proposes a new approach to model the temporal dynamics of a sequence of facial expressions. To this purpose, a sequence of Face Image Descriptors (FID) is regarded as the output of a Linear Time Invariant (LTI) system. The temporal dynamics of such sequence of descriptors are represented by means of a Hankel matrix. The paper presents different strategies to compute dynamics-based representation of a sequence of FID, and reports classification accuracy values of the proposed representations within different standard classification frameworks. The representations have been validated in two very challenging application domains: emotion recognition and pain detection. Experiments on two publicly available benchmarks and comparison with state-of-the-art approaches demonstrate that the dynamics-based FID representation attains competitive performance when off-the- shelf classification tools are adopted.File | Dimensione | Formato | |
---|---|---|---|
CVPRW_2015.pdf
Solo gestori archvio
Descrizione: Articolo principale
Tipologia:
Versione Editoriale
Dimensione
288.82 kB
Formato
Adobe PDF
|
288.82 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Presti_Using_Hankel_Matrices_2015_CVPR_paper.pdf
accesso aperto
Descrizione: workshop paper
Tipologia:
Post-print
Dimensione
308.24 kB
Formato
Adobe PDF
|
308.24 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.