hh.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Hernandez-Diaz, Kevin
Publications (3 of 3) Show all publications
Hernandez-Diaz, K., Alonso-Fernandez, F. & Bigun, J. (2019). Cross Spectral Periocular Matching using ResNet Features. In: : . Paper presented at 12th IAPR International Conference on Biometrics, Crete, Greece, June 4-7, 2019.
Open this publication in new window or tab >>Cross Spectral Periocular Matching using ResNet Features
2019 (English)Conference paper, Published paper (Refereed)
Abstract [en]

Periocular recognition has gained attention in the last years thanks to its high discrimination capabilities in less constraint scenarios than other ocular modalities. In this paper we propose a method for periocular verification under different light spectra using CNN features with the particularity that the network has not been trained for this purpose. We use a ResNet-101 pretrained model for the ImageNet Large Scale Visual Recognition Challenge to extract features from the IIITD Multispectral Periocular Database. At each layer the features are compared using χ 2 distance and cosine similitude to carry on verification between images, achieving an improvement in the EER and accuracy at 1% FAR of up to 63.13% and 24.79% in comparison to previous works that employ the same database. In addition to this, we train a neural network to match the best CNN feature layer vector from each spectrum. With this procedure, we achieve improvements of up to 65% (EER) and 87% (accuracy at 1% FAR) in cross-spectral verification with respect to previous studies.

National Category
Signal Processing
Identifiers
urn:nbn:se:hh:diva-40499 (URN)
Conference
12th IAPR International Conference on Biometrics, Crete, Greece, June 4-7, 2019
Funder
Swedish Research Council, 2016-03497Knowledge Foundation, SIDUS-AIRKnowledge Foundation, CAISR
Available from: 2019-09-04 Created: 2019-09-04 Last updated: 2019-10-11
Hernandez-Diaz, K., Alonso-Fernandez, F. & Bigun, J. (2019). Cross-Spectral Biometric Recognition with Pretrained CNNs as Generic Feature Extractors. In: : . Paper presented at Swedish Symposium on Image Analysis, SSBA, Gothenburg, Sweden, March 19-20, 2019.
Open this publication in new window or tab >>Cross-Spectral Biometric Recognition with Pretrained CNNs as Generic Feature Extractors
2019 (English)Conference paper, Published paper (Other academic)
Abstract [en]

Periocular recognition has gained attention in the last years thanks to its high discrimination capabilities in less constraint scenarios than face or iris. In this paper we propose a method for periocular verification under different light spectra using CNN features with the particularity that the network has not been trained for this purpose. We use a ResNet-101 pretrained model for the ImageNet Large Scale Visual Recognition Challenge to extract features from the IIITD Multispectral Periocular Database. At each layer the features are compared using χ 2 distance and cosine similitude to carry on verification between images, achieving an improvement in the EER and accuracy at 1% FAR of up to 63.13% and 24.79% in comparison to previous works that employ the same database. In addition to this, we train a neural network to match the best CNN feature layer vector from each spectrum. With this procedure, we achieve improvements of up to 65% (EER) and 87% (accuracy at 1% FAR) in cross-spectral verification with respect to previous studies.

National Category
Signal Processing
Identifiers
urn:nbn:se:hh:diva-40625 (URN)
Conference
Swedish Symposium on Image Analysis, SSBA, Gothenburg, Sweden, March 19-20, 2019
Available from: 2019-09-24 Created: 2019-09-24 Last updated: 2019-10-11
Hernandez-Diaz, K., Alonso-Fernandez, F. & Bigun, J. (2018). Periocular Recognition Using CNN Features Off-the-Shelf. In: 2018 International Conference of the Biometrics Special Interest Group (BIOSIG): . Paper presented at International Conference of the Biometrics Special Interest Group (BIOSIG), Darmstadt, Germany, Sept. 26-29, 2018. Piscataway, N.J.: IEEE
Open this publication in new window or tab >>Periocular Recognition Using CNN Features Off-the-Shelf
2018 (English)In: 2018 International Conference of the Biometrics Special Interest Group (BIOSIG), Piscataway, N.J.: IEEE, 2018Conference paper, Published paper (Refereed)
Abstract [en]

Periocular refers to the region around the eye, including sclera, eyelids, lashes, brows and skin. With a surprisingly high discrimination ability, it is the ocular modality requiring the least constrained acquisition. Here, we apply existing pre-trained architectures, proposed in the context of the ImageNet Large Scale Visual Recognition Challenge, to the task of periocular recognition. These have proven to be very successful for many other computer vision tasks apart from the detection and classification tasks for which they were designed. Experiments are done with a database of periocular images captured with a digital camera. We demonstrate that these offthe-shelf CNN features can effectively recognize individuals based on periocular images, despite being trained to classify generic objects. Compared against reference periocular features, they show an EER reduction of up to ~40%, with the fusion of CNN and traditional features providing additional improvements.

Place, publisher, year, edition, pages
Piscataway, N.J.: IEEE, 2018
Series
2018 International Conference of the Biometrics Special Interest Group (BIOSIG), ISSN 1617-5468 ; 2018
Keywords
Periocular recognition, deep learning, biometrics, Convolutional Neural Network
National Category
Signal Processing
Identifiers
urn:nbn:se:hh:diva-37704 (URN)10.23919/BIOSIG.2018.8553348 (DOI)978-3-88579-676-3 (ISBN)
Conference
International Conference of the Biometrics Special Interest Group (BIOSIG), Darmstadt, Germany, Sept. 26-29, 2018
Projects
SIDUS-AIR
Funder
Knowledge Foundation, SIDUS-AIRSwedish Research Council, 2016-03497Vinnova, 2018-00472Knowledge Foundation, CAISR
Available from: 2018-08-14 Created: 2018-08-14 Last updated: 2019-05-23Bibliographically approved
Organisations

Search in DiVA

Show all publications