Eye centre localisation: an unsupervised modular approach

Wenhao Zhang,Abdul Rehman Farooq,Melvyn Lionel Smith,Lyndon Neal Smith

doi:10.1108/sr-06-2015-0098

Wenhao Zhang, Abdul Rehman Farooq + Show 2 more

Open Access

https://doi.org/10.1108/sr-06-2015-0098

Copy DOI

Abstract

Purpose This paper aims to introduce an unsupervised modular approach for eye centre localisation in images and videos following a coarse-to-fine, global-to-regional scheme. The design of the algorithm aims at excellent accuracy, robustness and real-time performance for use in real-world applications. Design/methodology/approach A modular approach has been designed that makes use of isophote and gradient features to estimate eye centre locations. This approach embraces two main modalities that progressively reduce global facial features to local levels for more precise inspections. A novel selective oriented gradient (SOG) filter has been specifically designed to remove strong gradients from eyebrows, eye corners and self-shadows, which sabotage most eye centre localisation methods. The proposed algorithm, tested on the BioID database, has shown superior accuracy. Findings The eye centre localisation algorithm has been compared with 11 other methods on the BioID database and six other methods on the GI4E database. The proposed algorithm has outperformed all the other algorithms in comparison in terms of localisation accuracy while exhibiting excellent real-time performance. This method is also inherently robust against head poses, partial eye occlusions and shadows. Originality/value The eye centre localisation method uses two mutually complementary modalities as a novel, fast, accurate and robust approach. In addition, other than assisting eye centre localisation, the SOG filter is able to resolve general tasks regarding the detection of curved shapes. From an applied point of view, the proposed method has great potentials in benefiting a wide range of real-world human-computer interaction (HCI) applications.

Highlights

Introduction viEye centre localisation from images and videos has received a considerable amount of attention in the area of HCI through utilization of computer vision techniques
Apart from the studies reviewed we provide a detailed summary of 11 state-of-the-art methods in section 3 and a comparison with our eye centre localisation method
To resolve the challenges posed by strong shadows inside an eye region and sharp edges outside the eye region, we introduce a radius constraint and design a Selective Oriented Gradient (SOG) filter that can effectively deal with the circularity measure and automatically remove strong gradients from eyebrows and eyelids

Summary

Introduction

Eye centre localisation from images and videos has received a considerable amount of attention in the area of HCI through utilization of computer vision techniques. The ability to accurately localise eye centres can promise to bring significant benefits to a HCI system that is designed to observe its users and to capture user attentiveness. With the knowledge of user attention and predicted user intentions, a HCI system can make informed decisions and can respond to a user in a more intelligent and natural way. Compared to other facial cues of a particular user such as age and gender that remain unchanged during a HCI session, eye analysis provides a constant stream of information that can address the dynamic nature of HCI. Eye analysis excels in remote and contactless HCI, which provides an ideal channel for elderly people and those with motor disabilities to access HCI systems

Methods

Results

Conclusion