GUC Light Field Face and Iris Database

 

GUC Li ght F ield F ace and Iris D atabase (LiFFID ) are composed in the context of the EU FP7 FIDELITY project. The database collection was carried out during the period of November 2012 till Feb 2014 at Gjøvik University College (GUC), Norway. The database was collected using commercially available light field camera from Lytro. There are two kind of databases that are collected:

The GUC Light Field Face Database : This database comprises of 112 subjects whose faces are captured using both high resolution DSLR Camera and Lytro light field camera. The enrollment samples represent by the high quality images captured using DSLR camera while probe samples are represented by the Lytro light field camera. The probe samples are collected in three different protocols by simulating real life surveillance scenario. There are more than 1165 probe samples that represent the multiple subject that constitute for more than 2755 face samples.

Regarding the protocols for evaluation, baseline performance and more statistics refer the following publication.

The GUC Light Field Visible IRIS Database : This database is comprised of 55 subjects that can represent 110 unique eye pattern. There are 5 samples captured for each eye using Lytro light field camera by simulating real life scenario. For more information about protocols for evaluation, baseline performance and more statistics refer the following publication.

At present, we are in the process of obtaining the permission from Norsk samfunnsvitenskapelig datatjeneste AS (NSD) (http://www.nsd.uib.no/) , Norway in order make this database available to the public. This procedure is mandatory in Norway to respect the privacy issue.

Contact:
Dr. Raghavendra Ramachandra (raghavendra.ramachandra@ntnu.no)

Publications:

[1] R. Raghavendra, Kiran B. Raja, Christoph Busch, ‘Exploring the Usefulness of Light Field Cameras for Biometrics : An Empirical Study on Face and Iris Recognition’, IEEE Transactions on Information Forensics and Security (TIFS),  2016. (Accepted)