Near-infrared and visible-light periocular recognition with Gabor features using frequency-adaptive automatic eye detection

TL;DR

This paper introduces a robust periocular recognition system using Gabor features and a novel, training-free eye detection method based on complex symmetry filters, evaluated across multiple near-infrared and visible-light datasets.

Contribution

It presents a new, training-free eye detection approach and demonstrates its effectiveness within a periocular recognition system that benefits from multi-database evaluation and fusion with iris matchers.

Findings

High accuracy in near-infrared data

Good accuracy in visible-light data

Fusion improves recognition performance by over 20%

Abstract

Periocular recognition has gained attention recently due to demands of increased robustness of face or iris in less controlled scenarios. We present a new system for eye detection based on complex symmetry filters, which has the advantage of not needing training. Also, separability of the filters allows faster detection via one-dimensional convolutions. This system is used as input to a periocular algorithm based on retinotopic sampling grids and Gabor spectrum decomposition. The evaluation framework is composed of six databases acquired both with near-infrared and visible sensors. The experimental setup is complemented with four iris matchers, used for fusion experiments. The eye detection system presented shows very high accuracy with near-infrared data, and a reasonable good accuracy with one visible database. Regarding the periocular system, it exhibits great robustness to small errors in locating the eye centre, as well as to scale changes of the input image. The density of the sampling grid can also be reduced without sacrificing accuracy. Lastly, despite the poorer performance of the iris matchers with visible data, fusion with the periocular system can provide an improvement of more than 20%. The six databases used have been manually annotated, with the annotation made publicly available.