Unknown Face Presentation Attack Detection via Localised Learning of Multiple Kernels

TL;DR

This paper introduces a localised multiple kernel learning approach for face presentation attack detection that adapts to local data structures, improving detection of unknown attack types in zero-shot scenarios.

Contribution

It proposes a convex localised MKL algorithm with joint matrix-norm constraints, enhancing unknown attack detection by leveraging local data variability.

Findings

The method outperforms other techniques in detecting unseen attacks.

Theoretical analysis confirms good generalisation capabilities.

Experimental results demonstrate effectiveness on face PAD datasets.

Abstract

The paper studies face spoofing, a.k.a. presentation attack detection (PAD) in the demanding scenarios of unknown types of attack. While earlier studies have revealed the benefits of ensemble methods, and in particular, a multiple kernel learning approach to the problem, one limitation of such techniques is that they typically treat the entire observation space similarly and ignore any variability and local structure inherent to the data. This work studies this aspect of the face presentation attack detection problem in relation to multiple kernel learning in a one-class setting to benefit from intrinsic local structure in bona fide face samples. More concretely, inspired by the success of the one-class Fisher null formalism, we formulate a convex localised multiple kernel learning algorithm by imposing a joint matrix-norm constraint on the collection of local kernel weights and infer locally adaptive weights for zero-shot one-class unseen attack detection. We present a theoretical study of the proposed localised MKL algorithm using Rademacher complexities to characterise its generalisation capability and demonstrate the advantages of the proposed technique over some other options. An assessment of the proposed approach on general object image datasets illustrates its efficacy for abnormality and novelty detection while the results of the experiments on face PAD datasets verifies its potential in detecting unknown/unseen face presentation attacks.