A Full Transformer-based Framework for Automatic Pain Estimation using Videos

TL;DR

This paper introduces a comprehensive transformer-based framework for automatic pain estimation from videos, achieving state-of-the-art performance and demonstrating strong generalization across pain assessment tasks.

Contribution

It presents a novel full transformer-based framework with specialized transformer modules for improved pain estimation from video data.

Findings

Achieved state-of-the-art performance on BioVid database

Demonstrated high efficacy and efficiency in pain estimation

Showed strong generalization across primary pain assessment tasks

Abstract

The automatic estimation of pain is essential in designing an optimal pain management system offering reliable assessment and reducing the suffering of patients. In this study, we present a novel full transformer-based framework consisting of a Transformer in Transformer (TNT) model and a Transformer leveraging cross-attention and self-attention blocks. Elaborating on videos from the BioVid database, we demonstrate state-of-the-art performances, showing the efficacy, efficiency, and generalization capability across all the primary pain estimation tasks.