# FaceSleuth-R: Adaptive Orientation-Aware Attention for Robust Micro-Expression Recognition

arXiv: https://arxiv.org/abs/2506.02695 · HTML: https://mlsanity.com/paper/2506.02695/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 17 references that are in the map.
- First posted on arXiv 3 Jun 2025.

## Builds on (most cited first, 5 of 5)

- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Micron-BERT: BERT-based Facial Micro-Expression Recognition](https://mlsanity.com/paper/2304.03195/) (2023)
- [MMNet: Muscle motion-guided network for micro-expression recognition](https://mlsanity.com/paper/2201.05297/) (2022)
- [Feature Representation Learning with Adaptive Displacement Generation and Transformer Fusion for Micro-Expression Recognition](https://mlsanity.com/paper/2304.04420/) (2023)
- [A Novel Apex-Time Network for Cross-Dataset Micro-Expression Recognition](https://mlsanity.com/paper/1904.03699/) (2019)

## Abstract (arXiv)

Micro-expression recognition (MER) has achieved impressive accuracy in controlled laboratory settings. However, its real-world applicability faces a significant generalization cliff, severely hindering practical deployment due to poor performance on unseen data and susceptibility to domain shifts. Existing attention mechanisms often overfit to dataset-specific appearance cues or rely on fixed spatial priors, making them fragile in diverse environments. We posit that robust MER requires focusing on quasi-invariant motion orientations inherent to micro-expressions, rather than superficial pixel-level features. To this end, we introduce FaceSleuth-R, a framework centered on our novel Single-Orientation Attention (SOA) module. SOA is a lightweight, differentiable operator that enables the network to learn layer-specific optimal orientations, effectively guiding attention towards these robust motion cues. Through extensive experiments, we demonstrate that SOA consistently discovers a universal near-vertical motion prior across diverse datasets. More critically, FaceSleuth-R showcases superior generalization in rigorous Leave-One-Dataset-Out (LODO) protocols, significantly outperforming baselines and state-of-the-art methods when confronted with domain shifts. Furthermore, our approach establishes state-of-the-art results across several benchmarks. This work highlights adaptive orientation-aware attention as a key paradigm for developing truly generalized and high-performing MER systems.