Iterative Optimization of Pseudo Ground-Truth Face Image Quality Labels

TL;DR

This paper introduces an iterative optimization method that enhances face image quality scores by leveraging mated similarity scores, improving the performance of existing FIQA techniques in face recognition systems.

Contribution

It proposes a novel iterative optimization approach that refines face image quality labels by incorporating similarity score information, outperforming baseline FIQA methods.

Findings

Optimized quality scores outperform original FIQA scores.

Performance improves with more optimization iterations, peaking at ten.

Method is validated across three datasets and three FIQA techniques.

Abstract

While recent face recognition (FR) systems achieve excellent results in many deployment scenarios, their performance in challenging real-world settings is still under question. For this reason, face image quality assessment (FIQA) techniques aim to support FR systems, by providing them with sample quality information that can be used to reject poor quality data unsuitable for recognition purposes. Several groups of FIQA methods relying on different concepts have been proposed in the literature, all of which can be used for generating quality scores of facial images that can serve as pseudo ground-truth (quality) labels and can be exploited for training (regression-based) quality estimation models. Several FIQA appro\-aches show that a significant amount of sample-quality information can be extracted from mated similarity-score distributions generated with some face matcher. Based on this insight, we propose in this paper a quality label optimization approach, which incorporates sample-quality information from mated-pair similarities into quality predictions of existing off-the-shelf FIQA techniques. We evaluate the proposed approach using three state-of-the-art FIQA methods over three diverse datasets. The results of our experiments show that the proposed optimization procedure heavily depends on the number of executed optimization iterations. At ten iterations, the approach seems to perform the best, consistently outperforming the base quality scores of the three FIQA methods, chosen for the experiments.