FME Workshop: Pushing Boundaries in Temporal and Spatial Subtle Movement Analysis

Facial micro-expressions (MEs) are involuntary movements of the face that occur spontaneously when a person experiences an emotion but attempts to suppress or repress the facial expression, typically found in a high-stakes environments. MEs are very short, generally being no more than 500 milliseconds and the data used is often very challenging with the limited number of labelled ME samples. It is also near impossible to unify the standardisation of ME labelling for different annotators. This workshop aims to explore advanced techniques of micro-facial expression analysis using a multimodal approach. We expect new advancements in multimodal micro-expression approaches, using the usual vision and temporal images alongside other metadata. In addition, the rise in large language models and visual language models will further push the boundaries of analysis and overall performance.

ME Grand Challenge: Micro-Expression Visual-Question-Answering from Short to Long Videos

Facial micro-expressions (MEs) are involuntary movements of the face that occur spontaneously when a person experiences an emotion but attempts to suppress or repress the facial expression, typically found in a high-stakes environment. MEGC 2026 introduces two tasks:

• ME-VQA, a ME visual question answering task using short video clips; and
• ME-LVQA, a new micro-expression long video-question-answering task requiring participants to use long ME videos and provide natural-language answers. The competition leverages spontaneous ME datasets, provides baselines, submissions through Codabench, unified evaluation metrics, and aims to advance subtle facial behaviour understanding, temporal reasoning, and interpretable vision–language modelling.