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

Pain is a serious and costly issue globally, but to be treated, it must firstbe detected. Vision transformers are a top-performing architecture in computervision, with little research on their use for pain detection. In this paper, wepropose the first fully-attentive automated pain detection pipeline thatachieves state-of-the-art performance on binary pain detection from facialexpressions. The model is trained on the UNBC-McMaster dataset, after faces are3D-registered and rotated to the canonical frontal view. In our experiments weidentify important areas of the hyperparameter space and their interaction withvision and video vision transformers, obtaining 3 noteworthy models. We analysethe attention maps of one of our models, finding reasonable interpretations forits predictions. We also evaluate Mixup, an augmentation technique, andSharpness-Aware Minimization, an optimizer, with no success. Our presentedmodels, ViT-1 (F1 score 0.55 +- 0.15), ViViT-1 (F1 score 0.55 +- 0.13), andViViT-2 (F1 score 0.49 +- 0.04), all outperform earlier works, showing thepotential of vision transformers for pain detection. Code is available athttps://github.com/IPDTFE/ViT-McMaster

Fully-attentive and interpretable: vision and video vision transformers for pain detection