Expression Empowered ResiDen Network for Facial Action Unit Detection

TL;DR

This paper introduces ResiDen, a novel network combining residual connections and dense blocks, leveraging facial expression information to improve facial action unit detection in unconstrained settings, achieving state-of-the-art results.

Contribution

The paper proposes ResiDen, a new network architecture that integrates residual connections and dense blocks, and utilizes expression recognition features for enhanced AU detection.

Findings

Facial expression information improves AU detection accuracy.

ResiDen outperforms existing methods on EmotionNet and DISFA datasets.

Residual connections across dense blocks are beneficial for face analysis.

Abstract

The paper explores the topic of Facial Action Unit (FAU) detection in the wild. In particular, we are interested in answering the following questions: (1) how useful are residual connections across dense blocks for face analysis? (2) how useful is the information from a network trained for categorical Facial Expression Recognition (FER) for the task of FAU detection? The proposed network (ResiDen) exploits dense blocks along with residual connections and uses auxiliary information from a FER network. The experiments are performed on the EmotionNet and DISFA datasets. The experiments show the usefulness of facial expression information for AU detection. The proposed network achieves state-of-art results on the two databases. Analysis of the results for cross database protocol shows the effectiveness of the network.