Learn Convolutional Neural Network for Face Anti-Spoofing

Abstract

Though having achieved some progresses, the hand-crafted texture features, e.g., LBP [23], LBP-TOP [11] are still unable to capture the most discriminative cues between genuine and fake faces. In this paper, instead of designing feature by ourselves, we rely on the deep convolutional neural network (CNN) to learn features of high discriminative ability in a supervised manner. Combined with some data pre-processing, the face anti-spoofing performance improves drastically. In the experiments, over 70% relative decrease of Half Total Error Rate (HTER) is achieved on two challenging datasets, CASIA [36] and REPLAY-ATTACK [7] compared with the state-of-the-art. Meanwhile, the experimental results from inter-tests between two datasets indicates CNN can obtain features with better generalization ability. Moreover, the nets trained using combined data from two datasets have less biases between two datasets.

8 pages, 9 figures, 7 tables

Cited by in corpus (29)

• NAS-FAS: Static-Dynamic Central Difference Network Search for Face Anti-Spoofing
• Use of a Capsule Network to Detect Fake Images and Videos
• DRL-FAS: A Novel Framework Based on Deep Reinforcement Learning for Face Anti-Spoofing
• Learning Generalized Spoof Cues for Face Anti-spoofing
• Learning Generalizable and Identity-Discriminative Representations for Face Anti-Spoofing
• Single-Side Domain Generalization for Face Anti-Spoofing
• Cross-domain Face Presentation Attack Detection via Multi-domain Disentangled Representation Learning
• Face Anti-Spoofing Via Disentangled Representation Learning
• Deep Tree Learning for Zero-shot Face Anti-Spoofing
• Learning deep forest with multi-scale Local Binary Pattern features for face anti-spoofing
• Revisiting Pixel-Wise Supervision for Face Anti-Spoofing
• On Disentangling Spoof Trace for Generic Face Anti-Spoofing
• Physics-Guided Spoof Trace Disentanglement for Generic Face Anti-Spoofing
• Adaptive Normalized Representation Learning for Generalizable Face Anti-Spoofing
• A Compact Deep Learning Model for Face Spoofing Detection
• Dual Reweighting Domain Generalization for Face Presentation Attack Detection
• FaceSpoof Buster: a Presentation Attack Detector Based on Intrinsic Image Properties and Deep Learning
• Regularized Fine-grained Meta Face Anti-spoofing
• Two-stream Convolutional Networks for Multi-frame Face Anti-spoofing
• Attacking CNN-based anti-spoofing face authentication in the physical domain
• Uncertainty-Aware Physically-Guided Proxy Tasks for Unseen Domain Face Anti-spoofing
• Audio-replay attack detection countermeasures
• Suppressing Spoof-irrelevant Factors for Domain-agnostic Face Anti-spoofing
• PipeNet: Selective Modal Pipeline of Fusion Network for Multi-Modal Face Anti-Spoofing
• Spoof Face Detection Via Semi-Supervised Adversarial Training
• Use of in-the-wild images for anomaly detection in face anti-spoofing
• Replay Spoofing Countermeasure Using Autoencoder and Siamese Network on ASVspoof 2019 Challenge
• Camera Invariant Feature Learning for Generalized Face Anti-spoofing
• Unseen Face Presentation Attack Detection Using Class-Specific Sparse One-Class Multiple Kernel Fusion Regression