Improving Face Anti-Spoofing by 3D Virtual Synthesis

TL;DR

This paper introduces a 3D virtual synthesis method to generate large-scale spoof face data, improving face anti-spoofing performance without expensive data collection.

Contribution

It proposes a novel 3D virtual synthesis technique for creating diverse spoof data, enhancing deep learning-based face anti-spoofing methods.

Findings

Synthetic data significantly improves anti-spoofing accuracy

The method reduces reliance on costly real spoof data

Enhanced performance with data balancing strategy

Abstract

Face anti-spoofing is crucial for the security of face recognition systems. Learning based methods especially deep learning based methods need large-scale training samples to reduce overfitting. However, acquiring spoof data is very expensive since the live faces should be re-printed and re-captured in many views. In this paper, we present a method to synthesize virtual spoof data in 3D space to alleviate this problem. Specifically, we consider a printed photo as a flat surface and mesh it into a 3D object, which is then randomly bent and rotated in 3D space. Afterward, the transformed 3D photo is rendered through perspective projection as a virtual sample. The synthetic virtual samples can significantly boost the anti-spoofing performance when combined with a proposed data balancing strategy. Our promising results open up new possibilities for advancing face anti-spoofing using cheap and large-scale synthetic data.