Hyp-OC: Hyperbolic One Class Classification for Face Anti-Spoofing

TL;DR

This paper introduces a novel hyperbolic one-class classification framework for face anti-spoofing, effectively detecting spoof attacks without prior knowledge of spoof samples, and demonstrates superior performance on multiple benchmarks.

Contribution

It proposes the first hyperbolic embedding approach for one-class face anti-spoofing, with new loss functions and stabilization techniques, outperforming existing methods.

Findings

Outperforms state-of-the-art on five benchmark datasets

Effective in detecting unknown spoof attacks

Introduces hyperbolic space training techniques for face anti-spoofing

Abstract

Face recognition technology has become an integral part of modern security systems and user authentication processes. However, these systems are vulnerable to spoofing attacks and can easily be circumvented. Most prior research in face anti-spoofing (FAS) approaches it as a two-class classification task where models are trained on real samples and known spoof attacks and tested for detection performance on unknown spoof attacks. However, in practice, FAS should be treated as a one-class classification task where, while training, one cannot assume any knowledge regarding the spoof samples a priori. In this paper, we reformulate the face anti-spoofing task from a one-class perspective and propose a novel hyperbolic one-class classification framework. To train our network, we use a pseudo-negative class sampled from the Gaussian distribution with a weighted running mean and propose two novel loss functions: (1) Hyp-PC: Hyperbolic Pairwise Confusion loss, and (2) Hyp-CE: Hyperbolic Cross Entropy loss, which operate in the hyperbolic space. Additionally, we employ Euclidean feature clipping and gradient clipping to stabilize the training in the hyperbolic space. To the best of our knowledge, this is the first work extending hyperbolic embeddings for face anti-spoofing in a one-class manner. With extensive experiments on five benchmark datasets: Rose-Youtu, MSU-MFSD, CASIA-MFSD, Idiap Replay-Attack, and OULU-NPU, we demonstrate that our method significantly outperforms the state-of-the-art, achieving better spoof detection performance.