Causal Intervention for Subject-Deconfounded Facial Action Unit Recognition

TL;DR

This paper introduces a causal inference framework with a novel intervention module to improve subject-invariant facial action unit recognition, achieving state-of-the-art results on benchmark datasets.

Contribution

It formulates a causal model for AU recognition and proposes CIS, a causal intervention module that deconfounds subject bias, enhancing recognition accuracy.

Findings

CIS improves AU recognition performance.

CISNet achieves state-of-the-art results on BP4D and DISFA.

Causal intervention effectively reduces subject bias.

Abstract

Subject-invariant facial action unit (AU) recognition remains challenging for the reason that the data distribution varies among subjects. In this paper, we propose a causal inference framework for subject-invariant facial action unit recognition. To illustrate the causal effect existing in AU recognition task, we formulate the causalities among facial images, subjects, latent AU semantic relations, and estimated AU occurrence probabilities via a structural causal model. By constructing such a causal diagram, we clarify the causal effect among variables and propose a plug-in causal intervention module, CIS, to deconfound the confounder \emph{Subject} in the causal diagram. Extensive experiments conducted on two commonly used AU benchmark datasets, BP4D and DISFA, show the effectiveness of our CIS, and the model with CIS inserted, CISNet, has achieved state-of-the-art performance.