Exploring Spatial-Temporal Dynamics in Event-based Facial Micro-Expression Analysis
This work addresses the challenge of capturing fast facial movements for applications like human-robot interaction and driver monitoring, though it is incremental as it focuses on dataset creation and baseline evaluations.
The paper tackled the problem of analyzing subtle facial micro-expressions by introducing a new dataset with synchronized RGB and event camera recordings, achieving 51.23% accuracy in Action Unit classification using event data compared to 23.12% with RGB, and high-quality frame reconstruction with SSIM=0.8513 and PSNR=26.89 dB.
Micro-expression analysis has applications in domains such as Human-Robot Interaction and Driver Monitoring Systems. Accurately capturing subtle and fast facial movements remains difficult when relying solely on RGB cameras, due to limitations in temporal resolution and sensitivity to motion blur. Event cameras offer an alternative, with microsecond-level precision, high dynamic range, and low latency. However, public datasets featuring event-based recordings of Action Units are still scarce. In this work, we introduce a novel, preliminary multi-resolution and multi-modal micro-expression dataset recorded with synchronized RGB and event cameras under variable lighting conditions. Two baseline tasks are evaluated to explore the spatial-temporal dynamics of micro-expressions: Action Unit classification using Spiking Neural Networks (51.23\% accuracy with events vs. 23.12\% with RGB), and frame reconstruction using Conditional Variational Autoencoders, achieving SSIM = 0.8513 and PSNR = 26.89 dB with high-resolution event input. These promising results show that event-based data can be used for micro-expression recognition and frame reconstruction.