5 March 2015 A new engineering approach to reveal correlation of physiological change and spontaneous expression from video images
Author Affiliations +
Abstract
Spontaneous expression is associated with physiological states, i.e., heart rate, respiration, oxygen saturation (SpO2%), and heart rate variability (HRV). There have yet not sufficient efforts to explore correlation of physiological change and spontaneous expression. This study aims to study how spontaneous expression is associated with physiological changes with an approved protocol or through the videos provided from Denver Intensity of Spontaneous Facial Action Database. Not like a posed expression, motion artefact in spontaneous expression is one of evitable challenges to be overcome in the study. To obtain a physiological signs from a region of interest (ROI), a new engineering approach is being developed with an artefact-reduction method consolidated 3D active appearance model (AAM) based track, affine transformation based alignment with opto-physiological mode based imaging photoplethysmography. Also, a statistical association spaces is being used to interpret correlation of spontaneous expressions and physiological states including their probability densities by means of Gaussian Mixture Model. The present work is revealing a new avenue of study associations of spontaneous expressions and physiological states with its prospect of applications on physiological and psychological assessment.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Fenglei Yang, Sijung Hu, Xiaoyun Ma, Harnani Hassan, Dongqing Wei, "A new engineering approach to reveal correlation of physiological change and spontaneous expression from video images", Proc. SPIE 9315, Design and Quality for Biomedical Technologies VIII, 93150M (5 March 2015); doi: 10.1117/12.2077120; https://doi.org/10.1117/12.2077120
PROCEEDINGS
9 PAGES


SHARE
Back to Top