Grains of Saliency: Optimizing Saliency-based Training of Biometric Attack Detection Models

TL;DR

This paper investigates how different levels of human visual saliency granularity affect the training of biometric attack detection models, showing that simple saliency post-processing improves generalization across CNNs.

Contribution

It introduces an analysis of saliency granularity levels in training, highlighting their impact on model generalization in biometric attack detection.

Findings

Saliency post-processing enhances model generalization.

Different saliency granularities have varying impacts on performance.

Simple techniques can effectively leverage human saliency data.

Abstract

Incorporating human-perceptual intelligence into model training has shown to increase the generalization capability of models in several difficult biometric tasks, such as presentation attack detection (PAD) and detection of synthetic samples. After the initial collection phase, human visual saliency (e.g., eye-tracking data, or handwritten annotations) can be integrated into model training through attention mechanisms, augmented training samples, or through human perception-related components of loss functions. Despite their successes, a vital, but seemingly neglected, aspect of any saliency-based training is the level of salience granularity (e.g., bounding boxes, single saliency maps, or saliency aggregated from multiple subjects) necessary to find a balance between reaping the full benefits of human saliency and the cost of its collection. In this paper, we explore several different levels of salience granularity and demonstrate that increased generalization capabilities of PAD and synthetic face detection can be achieved by using simple yet effective saliency post-processing techniques across several different CNNs.