Recognition of 3D facial expression from posed data

Manar D. Samad; Khan M. Iftekharuddin

doi:10.1117/12.2015603

17 May 2013 Recognition of 3D facial expression from posed data

Manar D. Samad, Khan M. Iftekharuddin

Proceedings Volume 8738, Three-Dimensional Imaging, Visualization, and Display 2013; 87380X (2013) https://doi.org/10.1117/12.2015603
Event: SPIE Defense, Security, and Sensing, 2013, Baltimore, Maryland, United States

Abstract

Although recognition of facial expression in 3D facial images has been an active research area, most of the prior works are limited to using full frontal facial images. These techniques primarily project 3D facial image on 2D and manually select landmarks in 2D projection to extract relevant features. Face recognition in 2D images can be challenging due to unconstrained conditions such as head pose, occlusion, and resulting loss of data. Similarly, pose variation in 3D facial imaging can also result in loss of data. In most of the current 3D facial recognition works, when 3D posed face data are projected onto 2D, additional data loss may render 2D facial expression recognition even more challenging. In comparison, this work proposes novel feature extraction directly from the 3D facial posed images without the need of manual selection of landmarks or projection of images in 2D space. This feature is obtained as the angle between consecutive 3D normal vectors on the vertex points aligned either horizontally or vertically across the 3D facial image. Our facial expression recognition results show that the feature obtained from vertices aligned vertically across the face yields the best accuracy for classification with an average 87.8% area under the ROC. The results further suggest that the same feature outperforms its horizontal counterpart in recognizing facial expressions for pose variation between 35º - 50º with average accuracy of 80% - 60%, respectively.

Citation Download Citation

Manar D. Samad and Khan M. Iftekharuddin "Recognition of 3D facial expression from posed data", Proc. SPIE 8738, Three-Dimensional Imaging, Visualization, and Display 2013, 87380X (17 May 2013); https://doi.org/10.1117/12.2015603

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
8 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

3D image processing

Facial recognition systems

Feature extraction

3D modeling

Clouds

Sensors

Databases

Show All Keywords

Keywords/Phrases

Search In:

Publication Years