A Masked Face Classification Benchmark on Low-Resolution Surveillance Images

TL;DR

This paper introduces SF-MASK, a new low-resolution face mask dataset with synthetic images to improve mask classification, demonstrating better classifier performance compared to existing datasets.

Contribution

The creation of the SF-MASK dataset with real and synthetic low-resolution images for improved face mask classification is a novel contribution.

Findings

Classifiers perform better on SF-MASK than existing datasets.

Synthetic images enhance intra-class variance coverage.

The dataset includes challenging cases with badly worn masks.

Abstract

We propose a novel image dataset focused on tiny faces wearing face masks for mask classification purposes, dubbed Small Face MASK (SF-MASK), composed of a collection made from 20k low-resolution images exported from diverse and heterogeneous datasets, ranging from 7 x 7 to 64 x 64 pixel resolution. An accurate visualization of this collection, through counting grids, made it possible to highlight gaps in the variety of poses assumed by the heads of the pedestrians. In particular, faces filmed by very high cameras, in which the facial features appear strongly skewed, are absent. To address this structural deficiency, we produced a set of synthetic images which resulted in a satisfactory covering of the intra-class variance. Furthermore, a small subsample of 1701 images contains badly worn face masks, opening to multi-class classification challenges. Experiments on SF-MASK focus on face mask classification using several classifiers. Results show that the richness of SF-MASK (real + synthetic images) leads all of the tested classifiers to perform better than exploiting comparative face mask datasets, on a fixed 1077 images testing set. Dataset and evaluation code are publicly available here: https://github.com/HumaticsLAB/sf-mask