US Patent Patent # 9.317.779 was awarded on 04/19/2016.
Object recognition is a
well studied but extremely challenging
field. We developed a novel approach to feature
construction for object detection called
Evolution-COnstructed Features (ECO features).
Most current approaches rely on human experts to
construct features for object recognition. ECO
features are automatically constructed by uniquely
employing a standard genetic algorithm to discover
series of transforms that are highly
discriminative. Using ECO features provides
several advantages over other object detection
algorithms including: no need for a human expert
to build feature sets or tune their parameters,
ability to generate specialized feature sets for
different objects, no limitations to certain types
of image sources, and ability to find both global
and local feature types. We show in our
experiments that the ECO features compete well
against state-of-the-art object recognition
Lillywhite, Beau Tippetts, Meng Zhang
M. Zhang and D.J. Lee, “Global ECO-Feature for Object Classification,”
Lecture Notes in Computer Science (LNCS), International Symposium on Visual Computing (ISVC), Part II, LNCS 10073, pp. 281–290, Las Vegas, NV, U.S.A, December 12-14, 2016.
Zhang and D.J. Lee, “Efficient Training of Evolution-COnstructed
Features,” Lecture Notes in Computer Science (LNCS),
International Symposium on Visual Computing (ISVC), Part II, LNCS 9475,
p. 646-654, Las Vegas, NV, U.S.A., December 14-16, 2015.
Lillywhite, D.J. Lee, B.J. Tippetts, and J.K
Archibald, “A Feature Construction Method
for General Object Recognition,”
vol. 46/12, p. 3300-3314, December 2013.
Lillywhite, B.J. Tippetts, and D.J. Lee, “Self-Tuned
Evolution-COnstructed Features for General
Pattern Recognition, vol. 45/1, p. 241-251,
Lillywhite, D.J. Lee, and B.J. Tippetts, “Improving
Evolution-COnstructed Features Using
Speciation,” IEEE Workshop on
Applications of Computer Vision (WACV), 6
pages, Breckenridge, Colorado, USA, January
image to view.)