Abstract
This paper introduces an unsupervised modular approach for accurate and real-time eye center localization in images and videos, thus allowing a coarse-to-fine, global-to-regional scheme. The trajectories of eye centers in consecutive frames, i.e., gaze gestures, are further analyzed, recognized, and employed to boost the human-computer interaction (HCI) experience. This modular approach makes use of isophote and gradient features to estimate the eye center locations. A selective oriented gradient filter has been specifically designed to remove strong gradients from eyebrows, eye corners, and shadows, which sabotage most eye center localization methods. A real-world implementation utilizing these algorithms has been designed in the form of an interactive advertising billboard to demonstrate the effectiveness of our method for HCI. The eye center localization algorithm has been compared with 10 other algorithms on the BioID database and six other algorithms on the GI4E database. It outperforms all the other algorithms in comparison in terms of localization accuracy. Further tests on the extended Yale Face Database b and self-collected data have proved this algorithm to be robust against moderate head poses and poor illumination conditions. The interactive advertising billboard has manifested outstanding usability and effectiveness in our tests and shows great potential for benefiting a wide range of real-world HCI applications.
Highlights
Human computer interaction (HCI), as a cross-disciplinary area, has been under extensive study for nearly half a century
Three publicly available databases were tested in our experiments: the BioID database [27], the GI4E database [16] and the extended Yale Face Database b [28]
The BioID database is the most widely employed database for eye centre localisation studies since it contains an array of variations including illumination, face scale, moderate head pose and the presence of glasses; The GI4E database is known for containing images of 103 subjects with 12 different gaze directions; The extended Yale Face Database b is captured under extremely challenging lighting conditions and contains various head poses
Summary
Human computer interaction (HCI), as a cross-disciplinary area, has been under extensive study for nearly half a century. It has become central to information science theoretically and professionally, and has stepped into peoples’ lives, offering multiple types of communication channels, i.e. modalities. These modalities, independently or when combined [1], have played different assistive roles in their corresponding HCI applications [2]. Common studies regarding eye/gaze analysis include eye centre localisation and gaze tracking These two closely related research topics have comprised various paradigms where the data acquisition devices, the eye models, the features and the classifiers may differ from one another. This is evidenced by their contributions to face registration [3], health monitoring, user awareness/attention monitoring, marketing, advertising and other sectors
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
