Close
Metadata

%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2018/10.21.17.37
%2 sid.inpe.br/sibgrapi/2018/10.21.17.37.17
%A Bresan, Rodrigo,
%A Rezende, Edmar,
%A Beluzo, Carlos,
%A Carvalho, Tiago,
%@affiliation Federal Institute of Sao Paulo (IFSP)
%@affiliation University of Campinas (UNICAMP)
%@affiliation Federal Institute of Sao Paulo (IFSP)
%@affiliation Federal Institute of Sao Paulo (IFSP), University of Campinas (UNICAMP)
%T Presentation Attack Detection by a Combination of Intrinsic Image Properties and a CNN
%B Conference on Graphics, Patterns and Images, 31 (SIBGRAPI)
%D 2018
%E Ross, Arun,
%E Gastal, Eduardo S. L.,
%E Jorge, Joaquim A.,
%E Queiroz, Ricardo L. de,
%E Minetto, Rodrigo,
%E Sarkar, Sudeep,
%E Papa, João Paulo,
%E Oliveira, Manuel M.,
%E Arbeláez, Pablo,
%E Mery, Domingo,
%E Oliveira, Maria Cristina Ferreira de,
%E Spina, Thiago Vallin,
%E Mendes, Caroline Mazetto,
%E Costa, Henrique Sérgio Gutierrez,
%E Mejail, Marta Estela,
%E Geus, Klaus de,
%E Scheer, Sergio,
%S Proceedings
%8 Oct. 29 - Nov. 1, 2018
%J Porto Alegre
%I Sociedade Brasileira de Computação
%C Foz do Iguaçu, PR, Brazil
%K biometrics, presentation attack detection, PAD, intrinsic properties, anti-spoofing, facial recognition.
%X The usage of face recognition for biometric systems has become widely adopted, since it allows the usage of a trait that is accessible to most of the people. Despite important progress on the field of face recognition, there is still a lack of works whose focus consists on the detection of presentation attacks. Presentation attacks occur when an imposter presents a synthetic sample in order to impersonate a valid user. For face biometric systems, this kind of attack is performed using a photograph, by playing a video of the user (commonly known as replay attack) or by making usage of 3D masks. Hereby, we propose a low-cost solution to detect these kind of attacks without the need of extra hardware. Our hypothesis is based on the fact that, through the extraction of intrinsic image properties, such as depth, saliency and illumination, it is possible to distinguish between a real biometric sample and a synthetic one. Performed experiments show that the proposed method achieved HTER values of 41.64% and 3.88% in inter and intra protocols respectively, achieving near state-of-the-art results.
%@language en
%3 sibgrapi_correcoes (7).pdf


Close