English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

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

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