Smile Detection Using Multi-scale Gaussian Derivatives - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2013

Smile Detection Using Multi-scale Gaussian Derivatives

Abstract

In this paper we show that Multi-scale Gaussian Derivatives combined with Support Vector Machines provide an effective approach to facial expression analysis.The approach was tested on the GENKI and Cohn-Kanade dataset for detecting smiles. Our approach outperformed the current benchmark approach of using Gabor Energy Filters with Support Vector Machines.
Fichier principal
Vignette du fichier
ELSINA-25.pdf (1017.13 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-00807362 , version 1 (03-04-2013)

Identifiers

  • HAL Id : hal-00807362 , version 1

Cite

Varun Jain, James L. Crowley. Smile Detection Using Multi-scale Gaussian Derivatives. 12th WSEAS International Conference on Signal Processing, Robotics and Automation, Feb 2013, Cambridge, United Kingdom. ⟨hal-00807362⟩
605 View
492 Download

Share

Gmail Mastodon Facebook X LinkedIn More