04-10-2010, 10:59 AM
This article is presented by:
Marc Lalonde
David Byrns
Langis Gagnon
Normand Teasdale
Denis Laurendeau
ABSTRACT
The aim of this paper is to report about a first version of a real-time eye blink detector in the context of car driving simulation. This detector was designed within the COBVIS-D project, whose goal is to develop a simulation environment with a multi-sensor data acquisition and analysis system for driving performance assessment, cognitive load measure and training. Concretely, subjects are asked to sit in a driving simulator and react to realistic scenarios while being monitored by head and gaze trackers as well as monochrome video cameras. Their cognitive load will vary according to the degree of complexity of the driving task, and it can be assessed by analyzing and characterizing various physiological and physical data, among which are facial features. The analysis of facial features can be carried out within the framework of the Facial Action Coding System (FACS) that allows the decomposition of facial expressions in terms of facial feature displacements called Action Units (AU) . Literature review conducted for the COBVIS-D project has identified the FACS-based approach as the more appropriate to in-car driving situations. Workload or fatigue induces slight facial changes that cannot be detected with other methods. One can mention also that there are other physiological measures like electrocardiography and skin conductivity that might be better to characterize mental workload but these types of data were not available due to design constraints. According to the physiology literature, the main facial muscles that reflect mental effort are the lateral frontalis, the corrugator supercilii, orbicularis oris and levator palpebrae superioris . These muscles are responsible respectively for facial feature changes like eyebrow raiser, eyebrow frowning, lip suck, and eye blink.
For more information about this article,please follow the link:
http://googleurl?sa=t&source=web&cd=1&ve...1.153.4522%26rep%3Drep1%26type%3Dpdf&ei=0GSpTKWvLoWAvgPMmZzoDA&usg=AFQjCNGAKiDr0Ib6YbvA0c7XMtMIutUHdA
Marc Lalonde
David Byrns
Langis Gagnon
Normand Teasdale
Denis Laurendeau
Real-time eye blink detection with GPU-based SIFT tracking
ABSTRACT
This paper reports on the implementation of a GPUbased, real-time eye blink detector on very low contrast images acquired under near-infrared illumination. This detector is part of a multi-sensor data acquisition and analysis system for driver performance assessment and training. Eye blinks are detected inside regions of interest that are aligned with the subject’s eyes at initialization. Alignment is maintained through time by tracking SIFT feature points that are used to estimate the affine transformation between the initial face pose and the pose in subsequent frames. The GPU implementation of the SIFT feature point extraction algorithm ensures real-time processing. An eye blink detection rate of 97% is obtained on a video dataset of 33,000 frames showing 237 blinks from 22 subjects.
INTRODUCTIONThe aim of this paper is to report about a first version of a real-time eye blink detector in the context of car driving simulation. This detector was designed within the COBVIS-D project, whose goal is to develop a simulation environment with a multi-sensor data acquisition and analysis system for driving performance assessment, cognitive load measure and training. Concretely, subjects are asked to sit in a driving simulator and react to realistic scenarios while being monitored by head and gaze trackers as well as monochrome video cameras. Their cognitive load will vary according to the degree of complexity of the driving task, and it can be assessed by analyzing and characterizing various physiological and physical data, among which are facial features. The analysis of facial features can be carried out within the framework of the Facial Action Coding System (FACS) that allows the decomposition of facial expressions in terms of facial feature displacements called Action Units (AU) . Literature review conducted for the COBVIS-D project has identified the FACS-based approach as the more appropriate to in-car driving situations. Workload or fatigue induces slight facial changes that cannot be detected with other methods. One can mention also that there are other physiological measures like electrocardiography and skin conductivity that might be better to characterize mental workload but these types of data were not available due to design constraints. According to the physiology literature, the main facial muscles that reflect mental effort are the lateral frontalis, the corrugator supercilii, orbicularis oris and levator palpebrae superioris . These muscles are responsible respectively for facial feature changes like eyebrow raiser, eyebrow frowning, lip suck, and eye blink.
For more information about this article,please follow the link:
http://googleurl?sa=t&source=web&cd=1&ve...1.153.4522%26rep%3Drep1%26type%3Dpdf&ei=0GSpTKWvLoWAvgPMmZzoDA&usg=AFQjCNGAKiDr0Ib6YbvA0c7XMtMIutUHdA