Liveness score-based regression neural networks for face anti-spoofing

TL;DR

This paper introduces a novel liveness score-based regression neural network for face anti-spoofing that reduces dependency on third-party networks and user-defined labels, achieving superior performance on multiple benchmarks.

Contribution

It proposes a new pseudo-discretized label encoding and an expected liveness score regression approach for improved face anti-spoofing accuracy.

Findings

Outperforms previous methods on four benchmarks

Effective in intra- and cross-dataset tests

Reduces reliance on third-party networks and user labels

Abstract

Previous anti-spoofing methods have used either pseudo maps or user-defined labels, and the performance of each approach depends on the accuracy of the third party networks generating pseudo maps and the way in which the users define the labels. In this paper, we propose a liveness score-based regression network for overcoming the dependency on third party networks and users. First, we introduce a new labeling technique, called pseudo-discretized label encoding for generating discretized labels indicating the amount of information related to real images. Secondly, we suggest the expected liveness score based on a regression network for training the difference between the proposed supervision and the expected liveness score. Finally, extensive experiments were conducted on four face anti-spoofing benchmarks to verify our proposed method on both intra-and cross-dataset tests. The experimental results show our approach outperforms previous methods.