Gender-From-Iris or Gender-From-Mascara?

Andrey Kuehlkamp, Benedict Becker, Kevin Bowyer

Abstract:

Predicting a person’s gender based on the iris texture
has been explored by several researchers. This paper con-
siders several dimensions of experimental work on this
problem, including person-disjoint train and test, and the
effect of cosmetics on eyelash occlusion and imperfect seg-
mentation. We also consider the use of multi-layer percep-
tron and convolutional neural networks as classifiers, com-
paring the use of data-driven and hand-crafted features.
Our results suggest that the gender-from-iris problem is
more difficult than has so far been appreciated. Estimating
accuracy using a mean of N person-disjoint train and test
partitions, and considering the effect of makeup - a combi-
nation of experimental conditions not present in any previ-
ous work - we find a much weaker ability to predict gender-
from-iris texture than has been suggested in previous work.