Expression Empowered ResiDen Network for Facial Action Unit Detection
This work addresses facial analysis for applications like human-computer interaction, but it is incremental as it builds on existing methods with hybrid techniques.
The paper tackled Facial Action Unit detection in the wild by proposing the ResiDen network, which uses residual connections across dense blocks and auxiliary information from a Facial Expression Recognition network, achieving state-of-the-art results on EmotionNet and DISFA datasets.
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.