Teacher-Student Training and Triplet Loss to Reduce the Effect of Drastic Face Occlusion

TL;DR

This paper explores knowledge distillation techniques, including a novel triplet loss-based method, to improve face recognition under strong occlusion scenarios like VR headsets and masks, demonstrating consistent performance gains.

Contribution

Introduces a new triplet loss-based knowledge distillation method for face recognition tasks under occlusion, enhancing model performance across various neural architectures and recognition tasks.

Findings

Triplet loss-based distillation outperforms traditional methods.

Combining multiple distillation approaches yields further improvements.

Models show consistent gains across different neural architectures and recognition tasks.

Abstract

We study a series of recognition tasks in two realistic scenarios requiring the analysis of faces under strong occlusion. On the one hand, we aim to recognize facial expressions of people wearing Virtual Reality (VR) headsets. On the other hand, we aim to estimate the age and identify the gender of people wearing surgical masks. For all these tasks, the common ground is that half of the face is occluded. In this challenging setting, we show that convolutional neural networks (CNNs) trained on fully-visible faces exhibit very low performance levels. While fine-tuning the deep learning models on occluded faces is extremely useful, we show that additional performance gains can be obtained by distilling knowledge from models trained on fully-visible faces. To this end, we study two knowledge distillation methods, one based on teacher-student training and one based on triplet loss. Our main contribution consists in a novel approach for knowledge distillation based on triplet loss, which generalizes across models and tasks. Furthermore, we consider combining distilled models learned through conventional teacher-student training or through our novel teacher-student training based on triplet loss. We provide empirical evidence showing that, in most cases, both individual and combined knowledge distillation methods bring statistically significant performance improvements. We conduct experiments with three different neural models (VGG-f, VGG-face, ResNet-50) on various tasks (facial expression recognition, gender recognition, age estimation), showing consistent improvements regardless of the model or task.