Title

Technical Papers Session III: Optical assistance for Motor Neuron Disease (MND) patients using real-time eye tracking

Abstract/Description

Face detection and eyes extraction has an important role in many applications such as face recognition, facial expression analysis, security login etc. The system is designed to develop an eye-motion based communication system for patients suffering from motor neuron disease (MND) to contact with care takers whenever they want. In this paper, we proposed a method to efficiently track eye gaze detection in real time from video acquired by a web camera. The camera is stationary with respect to the head. Proposed system is mainly divided into three steps; face detection, eye-ball(pupil) detection and finally eye gaze detection. Using predefined facial landmarks face is detected in real time. Then in the second step face point that were extracted in phase one helps to detect eye point and using circular Hough Transform eye pupil was accurately detected. Finally, in the last stage eye gaze was calculated using gaze ratio formula. The experiments are conducted on people from different age groups and a high accuracy rate is achieved.

Location

Room C9 (Aman Tower, 3rd floor)

Session Theme

Technical Papers Session III - Computer Vision

Session Type

Parallel Technical Session

Session Chair

Dr. Asim Ur Rehman Khan

Start Date

16-11-2019 2:30 PM

End Date

16-11-2019 2:50 PM

Share

COinS
 
Nov 16th, 2:30 PM Nov 16th, 2:50 PM

Technical Papers Session III: Optical assistance for Motor Neuron Disease (MND) patients using real-time eye tracking

Room C9 (Aman Tower, 3rd floor)

Face detection and eyes extraction has an important role in many applications such as face recognition, facial expression analysis, security login etc. The system is designed to develop an eye-motion based communication system for patients suffering from motor neuron disease (MND) to contact with care takers whenever they want. In this paper, we proposed a method to efficiently track eye gaze detection in real time from video acquired by a web camera. The camera is stationary with respect to the head. Proposed system is mainly divided into three steps; face detection, eye-ball(pupil) detection and finally eye gaze detection. Using predefined facial landmarks face is detected in real time. Then in the second step face point that were extracted in phase one helps to detect eye point and using circular Hough Transform eye pupil was accurately detected. Finally, in the last stage eye gaze was calculated using gaze ratio formula. The experiments are conducted on people from different age groups and a high accuracy rate is achieved.