The field of facial expression recognition is moving towards more accurate and robust methods for recognizing subtle facial cues, such as micro-expressions. Researchers are exploring new representations and architectures that can better capture the temporal and dynamic nature of facial expressions. Notably, the use of dynamic images and phase-aware models is becoming increasingly popular. These advancements have significant implications for applications in psychology, security, and human-computer interaction. Noteworthy papers include:

• Adaptive Fusion Network with Temporal-Ranked and Motion-Intensity Dynamic Images for Micro-expression Recognition, which proposes a novel method for micro-expression recognition that achieves state-of-the-art results on several benchmark datasets.
• DIANet: A Phase-Aware Dual-Stream Network for Micro-Expression Recognition via Dynamic Images, which introduces a dual-stream framework that leverages phase-aware dynamic images to improve micro-expression recognition.
• ExpressNet-MoE: A Hybrid Deep Neural Network for Emotion Recognition, which proposes a hybrid model that blends convolutional neural networks and mixture of experts framework to improve emotion recognition accuracy.
• High Semantic Features for the Continual Learning of Complex Emotions: a Lightweight Solution, which presents a lightweight solution for incremental learning of complex emotions using high semantic features.