Face liveness detection: fusing colour texture feature and deep feature

Abstract

The identification which uses biological characteristics has been a current top in the recentpast. However, numerous spoofing skills occur with the rising prosperity ofadvance recognition technology, especially in the detection and recognition of aface. In allusion to the problem above, more robust and accurate face spoofingdetection schemes have been put forward. Convolutional neural networks (CNNs)have demonstrated extraordinary success in face liveness detection recently. Inthis study, an effective face anti-spoofing detection method based on CNN androtation invariant local binary patterns (RI-LBP) has been proposed. First, theauthors use CNN to extract deep features and use RI-LBP to extract colourtexture features. In addition, the principal component analysis approach isemployed to decrease the dimensions of deep characteristic. Moreover, twodifferent features are fused before applying to support vector machine (SVM).Finally, the SVM classifier is adopted to identify genuine faces from fakefaces. They have conducted extensive experiments to obtain a scheme of bettergeneralisation capability for face anti-spoofing detection. The analysis resultsindicate that the proposed approach implements great generalisation capabilityover other state-of-the-art approaches within the intra-databases andcross-databases.