IST Lenslet Light Field Face Spoofing Database (LLFFSD)
The IST Lenslet Light Field Face Spoofing Database (IST LLFFSD) is made available to address the lack of publicly available light field face resources to be used as the basis for the design, testing and validation of novel light field imaging based face spoofing detection systems.
The database consists of 50 subjects, with images captured by a Lytro ILLUM light field camera, at the Multimedia signal Processing Group of Instituto de Telecomunicações – Instituto Superior Técnico, Lisbon, Portugal. Two separate sessions were performed for each subject with a temporal separation between 1 and 6 months. The database includes 2 bona fide image shots per person one in each of the sessions, in a total of 100 bona fide image samples, plus six types of presentation attacks: printed paper, wrapped printed paper, laptop, tablet and two different mobile phones, in a total of 600 presentation attack images .
The files composing the database are organized according to a hierarchical structure. The root level of the hierarchy includes the folders for the bona fide images and each of the six attack types. Each of these folders contains three subfolders: “LFR”, “2D rendered images” and “landmark” (providing the location of faces in the images).
The naming convention for the LLFFSD database is ii_ttt_ss_n where:
· “ii” refers to the type of image, notably “LF” (Light Field) or “2D” (2 Dimensional);
· “ttt” is a three letter acronym identifying the nature of the images, notably “Bon” (Bona fide), “Lap” (Laptop), “Tab” (Tablet), “Mo1” (Mobile 1), “Mo2” (Mobile 2), “Pap” (paper), and “Wpa” (Wrapped paper);
· “ss” is a two digit integer uniquely identifying the subject, starting from 01.
· “n” is a digit indicating the acquisition session number, notably “1” or “2”;
· The IST Lenslet Light Field Face Spoofing Database is the first database to include light field images to test presentation attack detection systems. Besides the raw light field images, the LLFFSD database also includes sample 2D rendered images.
· The database can be downloaded from the links below (the size of each part is around 4GB).
· A password for decrypting the compressed Zip files will be provided after receiving the duly signed license agreement (see above).
How to reference the database:
· A. Sepas-Moghaddam, L. Malhadas, P.L. Correia, F. Pereira, “Face Spoofing Detection using a Light Field Imaging Framework”, IET Biometrics, Vol. 7, No.1, pp. 39-48, Jan. 2018
· A. Sepas-Moghaddam, F. Pereira, P.L. Correia, "Light Field based Face Presentation Attack Detection: Reviewing, Benchmarking and One Step Further", Vol. 13, No. 7, pp. 1696 - 1709, July 2018
Feedback is welcome. Please provide your comments and suggestions by e-mail.