Micro-expression is the involuntary emotion of the human that reflects the genuine feelings that cannot be hidden. Micro-expression is exhibited by facial expressions that last for a short duration and have very low intensity. Because of these reasons, micro-expression recognition is a challenging task. Recent research on the application of 3D convolutional neural networks (CNNs) has gained much popularity for video-based micro-expression analysis. For this purpose, both spatial as well as temporal features are of great importance to achieve high accuracies. The real possibly suppressed emotions of a person are valuable information for a variety of applications, such as in security, psychology, neuroscience, medicine and many other disciplines. This paper proposes a 3D CNN model architecture which is able to extract spatial and temporal features simultaneously. Thereby, the selection of the frame sequence plays a crucial role, since the emotions are only distinctive in a subset of the frames. Thus, we employ a novel pre-processing technique to select the Apex frame sequence from the entire video, where the timestamp of the most pronounced emotion is centered within this sequence. After an extensive evaluation including many experiments, the results show that the train–test split evaluation is biased toward a particular split and cannot be recommended in case of small and imbalanced datasets. Instead, a stratified K-fold evaluation technique is utilized to evaluate the model, which proves to be much more appropriate when using the three benchmark datasets CASME II, SMIC, and SAMM. Moreover, intra-dataset as well as cross-dataset evaluations were conducted in a total of eight different scenarios. For comparison purposes, two networks from the state of the art were reimplemented and compared with the presented architecture. In stratified K-fold evaluation, our proposed model outperforms both reimplemented state-of-the-art methods in seven out of eight evaluation scenarios.
30 September 2026 in Proceedings of the Institution of Civil Engineers - Municipal Engineer
Urban mobility systems continued to be strained due to traffic disturbances, including congestion, accidents, and emissions, resulting in significant financial, economic, and safety losses. Especially for stochastic disruptions or unforeseen events, such as traffic anomalies caused by various traffic disturbances, accidents and severe weather conditions, traditional traffic management systems and signal controls struggle to cope. To overcome these limitations, the transport ecosystem should transform the infrastructure to involve more intelligent, predictive, and cooperative participants. This work highlights several research efforts and complementary modules developed at TH Aschaffenburg over the past few years. An explainable and accident-aware congestion prediction framework is designed using a Bayesian network, followed by a multi-class fog modelling for weather-aware, robust perception. Moreover, ongoing research on the development and integration of reinforcement learning-based adaptive traffic signal control for traffic optimisation under uncertainty, a vehicle-to-everything-ready architecture for cooperative vehicle-infrastructure interaction, and the development of a digital twin framework for achieving a robust and reliable strategy through rigorous testing and validation is highlighted. Although these modules are studied individually, TH Aschaffenburg’s LiDAR-equipped City Mobility Living Lab facilitates the creation of an artificial intelligence-driven platform to unify, validate, and extend this research within real-world settings.