Micro Gesture Recognition

Micro-gesture recognition focuses on automatically identifying subtle, unintentional body movements, often linked to emotional states, which differ significantly from larger, intentional gestures. Current research emphasizes improving classification accuracy using techniques like prototype learning and contrastive learning, often incorporating multimodal data (e.g., video and text) and exploring efficient model architectures such as spiking neural networks for low-power applications. This field is significant for advancing artificial emotional intelligence and human-computer interaction, with applications ranging from improved emotion understanding in healthcare to more natural and intuitive interfaces in wearable technology.