Action Units Recognition by Pairwise Deep Architecture

TL;DR

This paper introduces a pairwise deep learning architecture for automatic Action Units recognition, significantly improving accuracy in a challenging in-the-wild dataset compared to baseline methods.

Contribution

The novel pairwise deep architecture effectively addresses label inconsistency in AU recognition, achieving a substantial performance boost in the ABAW competition.

Findings

Validation score improved from 0.31 to 0.67

Demonstrates effectiveness of pairwise approach in real-world AU recognition

Outperforms baseline in competition setting

Abstract

In this paper, we propose a new automatic Action Units (AUs) recognition method used in a competition, Affective Behavior Analysis in-the-wild (ABAW). Our method tackles a problem of AUs label inconsistency among subjects by using pairwise deep architecture. While the baseline score is 0.31, our method achieved 0.67 in validation dataset of the competition.