TY - GEN
T1 - Evaluation of Feature Channels for Correlation-Filter-Based Visual Object Tracking in Infrared Spectrum
AU - Gundogdu, Erhan
AU - Koc, Aykut
AU - Solmaz, Berkan
AU - Hammoud, Riad I.
AU - Alatan, A. Aydin
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/12/16
Y1 - 2016/12/16
N2 - Correlation filters for visual object tracking in visible imagery has been well-studied. Most of the correlation-filterbased methods use either raw image intensities or feature maps of gradient orientations or color channels. However, well-known features designed for visible spectrum may not be ideal for infrared object tracking, since infrared and visible spectra have dissimilar characteristics in general. We assess the performance of two state-of-the-art correlationfilter-based object tracking methods on Linköping Thermal InfraRed (LTIR) dataset of medium wave and longwave infrared videos, using deep convolutional neural networks (CNN) features as well as other traditional hand-crafted descriptors. The deep CNN features are trained on an infrared dataset consisting of 16K objects for a supervised classification task. The highest performance in terms of the overlap metric is achieved when these deep CNN features are utilized in a correlation-filter-based tracker.
AB - Correlation filters for visual object tracking in visible imagery has been well-studied. Most of the correlation-filterbased methods use either raw image intensities or feature maps of gradient orientations or color channels. However, well-known features designed for visible spectrum may not be ideal for infrared object tracking, since infrared and visible spectra have dissimilar characteristics in general. We assess the performance of two state-of-the-art correlationfilter-based object tracking methods on Linköping Thermal InfraRed (LTIR) dataset of medium wave and longwave infrared videos, using deep convolutional neural networks (CNN) features as well as other traditional hand-crafted descriptors. The deep CNN features are trained on an infrared dataset consisting of 16K objects for a supervised classification task. The highest performance in terms of the overlap metric is achieved when these deep CNN features are utilized in a correlation-filter-based tracker.
UR - http://www.scopus.com/inward/record.url?scp=85010209420&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85010209420&partnerID=8YFLogxK
U2 - 10.1109/CVPRW.2016.43
DO - 10.1109/CVPRW.2016.43
M3 - Conference contribution
AN - SCOPUS:85010209420
T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
SP - 290
EP - 298
BT - Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2016
PB - IEEE Computer Society
T2 - 29th IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2016
Y2 - 26 June 2016 through 1 July 2016
ER -