Spoofing in 2D face recognition with 3D masks and anti-spoofing with Kinect

Abstract

The problem of detecting face spoofing attacks (presentation attacks) has recently gained a well-deserved popularity. Mainly focusing on 2D attacks forged by displaying printed photos or replaying recorded videos on mobile devices, a significant portion of these studies ground their arguments on the flatness of the spoofing material in front of the sensor. In this paper, we inspect the spoofing potential of subject-specific 3D facial masks for 2D face recognition. Additionally, we analyze Local Binary Patterns based countermeasures using both color and depth data, obtained by Kinect. For this purpose, we introduce the 3D Mask Attack Database (3DMAD), the first publicly available 3D spoofing database, recorded with a low-cost depth camera. Extensive experiments on 3DMAD show that easily attainable facial masks can pose a serious threat to 2D face recognition systems and LBP is a powerful weapon to eliminate it.

What You Can't See Can Help You - Extended-Range Imaging for 3D-Mask Presentation Attack Detection

“…Typical face-recognition (FR) methods are highly susceptible to presentation attacks (PA), also commonly called spoof attacks [5], [11]. The term 'presentation attacks' covers both impersonation as well as concealment attacks [6].…”

RGB-D datasets using microsoft kinect or similar sensors: a survey

“…For more comprehensive comparisons, besides these 20 14 , from top left to bottom right is the chronological order from the video. The curves under the images are the accelerometer data at 50 Hz of devices attached to the knife, the mixing spoon, the small spoon, the peeler, the glass, the oil bottle, and the pepper dispenser mentioned datasets above, another 26 extra RGB-D datasets for different applications are also added into the tables: Birmingham University Objects, Category Modeling RGB-D [104], Cornell Activity [47,92], Cornell RGB-D [48], DGait [12], Daily Activities with occlusions [1], Heidelberg University Scenes [63], Microsoft 7-scenes [78], MobileRGBD [96], MPII Multi-Kinect [93], MSR Action3D Dataset [97], MSR 3D Online Action [103], MSRGesture3D [50], DAFT [31], Paper Kinect [70], RGBD-HuDaAct [68], Stanford Scene Object [44], Stanford 3D Scene [105], Sun3D [101], SUN RGB-D [82], TST Fall Detection [28], UTD-MHAD [14], Vienna University Technology Object [2], Willow Garage [99], Workout SU-10 exercise [67] and 3D-Mask [24]. In addition, we name those datasets without original names by means of creation place or applications.…”

A Novel Face Recognition Algorithm based on the Deep Convolution Neural Network and Key Points Detection Jointed Local Binary Pattern Methodology

“…2, we demonstrate the face key points adopted by our algorithm. In addition, we compare our method with the algorithms proposed in [28][29][30].…”

“…2. The face key points adopted by our algorithm 85.5% B [29] 86.2% A- [30] 87.4% A-possible to the original input for the target feature mapping to the input space. Generally, the deep neural network could be categorized as the follows.…”