Databases and Competitions

Databases

  • Ana F. Sequeira, Lulu Chen, Peter Wild, Petru Radu, James Ferryman. Cross-Eyed - Reading Cross-Spectrum Iris/Periocular Dataset.

Database: http://crosseyed.eu/

The Reading Cross-Spectrum Iris/Periocular Dataset (Cross-Eyed) was used as the benchmark dataset for the 1st Cross-Spectrum Iris/Periocular Competition. The Cross-Eyed dataset is composed by both VW/RGB and NIR images captured with a custom developed dual spectrum imaging sensor, which acquires NIR and VW images synchronously. The images were captured from 120 volunteers of several nationalities, ethnicities and eye/skin colours, which contributed with partial face images in each spectrum both the iris and periocular images were obtained by manually cropping the face images.

  • Juliano Murari and Ana F. Sequeira, MobBIOfake - an iris database with printed images captured with a handheld device.

Database: http://mobilive2014.inescporto.pt/

The MobBIOfake database was constructed upon the set of iris images from the MobBIO Multimodal Database [João C. Monteiro, Hélder P. Oliveira, Ana Rebelo and Ana F. Sequeira. MobBIO 2013: 1st Biometric Recognition with Portable Devices]. The MobBIOfake is composed of a subset of 800 iris images from MobBIO and 800 printed fake copies, in a total of 1600 iris images. The fake samples were obtained from printed images of the original ones captured with the same handheld device and in similar conditions.

  • João C. Monteiro, Hélder P. Oliveira, Ana Rebelo and Ana F. Sequeira, MobBIO 2013 - Biometric Recognition with Portable Devices Competition.

Database: http://www.fe.up.pt/~mobBIO2013/

The MobBIO Multimodal Database comprises the biometric data from 105 volunteers. Each individual provided samples of face, iris and voice. The equipment used for the samples acquisition was an Asus Transformer Pad TF 300T, with Android version 4.1.1. The device has two cameras, one frontal and one back camera. The camera used was the back camera, version TF300T-000128, with 8 MP of resolution and autofocus.

Organisation of Biometric Competitions

  • Cross-Eyed 2017 - 2nd Cross-Spectrum Iris/Periocular Recognition Competition (January 2017 - April 2017)

Website: http://crosseyed.eu/

Organising committee: Ana F. Sequeira, Lulu Chen, James Ferryman

Embraced by IJCB2017 - IEEE International Joint Conference on Biometrics: Denver, USA, 1-4 October 2017.

  • Cross-Eyed 2016 - 1st Cross-Spectrum Iris/Periocular Recognition Competition (January 2016 - April 2016)

Website: http://crosseyed.eu/

Organising committee: Ana F. Sequeira, Lulu Chen, James Ferryman

Embraced by BTAS2016 - IEEE International Conference on Biometrics: Theory Applications and Systems, Niagara Falls, USA, September, 2016.

The Cross-Eyed competition was organized in order to record recent advances in iris and periocular recognition across spectra. This competition covered two tasks: periocular and iris recognition and in both the main challenge was to perform comparison between NIR and VW images. Besides the importance of releasing new datasets to the research community, this competition aimed at recording advances in cross-spectrum iris/periocular recognition and to connect different research groups working on this topic.

  • MobILive 2014 – 1st Mobile Iris Liveness Detection Competition, (December 2013 - April 2014)

Website: http://mobilive2014.inescporto.pt/

Organising committee: Ana F. Sequeira, João C. Monteiro, Hélder P. Oliveira and Jaime S. Cardoso.

Embraced by IJCB2014 - IEEE Int. Joint Conference on Biometrics (IJCB), Clearwater, Florida, USA, 30 September – 02 October 2014.

The context of this competition is to perform iris liveness detection in mobile applications as an anti-spoofing measure (a step prior to the iris recognition process). The problem is, in fact, a two-class classification problem in which is intended to distinguish between real iris images and fake iris images. This competition covers a specific kind of attack (spoofing attack) in which printed images from an authorized user are presented to the sensor by a non-authorized user in order to obtain access.

Organising committee: Ana F. Sequeira, João C. Monteiro, Hélder P. Oliveira and Ana Rebelo.

Embraced by ICIAR2013 - "10th International Conference on Image Analysis and Recognition", Póvoa de Varzim, Portugal, 2013.

The main goal of the competition was to compare different methodologies for biometric recognition using data acquired with common portable devices, such as smartphones or tablets. The ambition of the competition is to become the reference event for academic and industrial researchers, especially in the area of security and personal identification. Establishing a public access multimodal biometric database represent an important contribution to the competition to the scientific community.