# Hierarchical Granularity Alignment and State Space Modeling for Robust Multimodal AU Detection in the Wild

arXiv: https://arxiv.org/abs/2603.11306 · HTML: https://mlsanity.com/paper/2603.11306/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 33 references that are in the map.
- First posted on arXiv 11 Mar 2026.

## Builds on (most cited first, 21 of 21)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model](https://mlsanity.com/paper/2401.09417/) (2024)
- [Expression, Affect, Action Unit Recognition: Aff-Wild2, Multi-Task Learning and ArcFace](https://mlsanity.com/paper/1910.04855/) (2019)
- [Deep Affect Prediction in-the-wild: Aff-Wild Database and Challenge, Deep Architectures, and Beyond](https://mlsanity.com/paper/1804.10938/) (2018)
- [Analysing Affective Behavior in the First ABAW 2020 Competition](https://mlsanity.com/paper/2001.11409/) (2020)
- [Affect Analysis in-the-wild: Valence-Arousal, Expressions, Action Units and a Unified Framework](https://mlsanity.com/paper/2103.15792/) (2021)
- [Distribution Matching for Heterogeneous Multi-Task Learning: a Large-scale Face Study](https://mlsanity.com/paper/2105.03790/) (2021)
- [Analysing Affective Behavior in the second ABAW2 Competition](https://mlsanity.com/paper/2106.15318/) (2021)
- [ABAW: Valence-Arousal Estimation, Expression Recognition, Action Unit Detection & Multi-Task Learning Challenges](https://mlsanity.com/paper/2202.10659/) (2022)
- [Face Behavior a la carte: Expressions, Affect and Action Units in a Single Network](https://mlsanity.com/paper/1910.11111/) (2019)
- [ABAW: Learning from Synthetic Data & Multi-Task Learning Challenges](https://mlsanity.com/paper/2207.01138/) (2022)
- [Learning Multi-dimensional Edge Feature-based AU Relation Graph for Facial Action Unit Recognition](https://mlsanity.com/paper/2205.01782/) (2022)
- [The 6th Affective Behavior Analysis in-the-wild (ABAW) Competition](https://mlsanity.com/paper/2402.19344/) (2024)
- [Distribution Matching for Multi-Task Learning of Classification Tasks: a Large-Scale Study on Faces & Beyond](https://mlsanity.com/paper/2401.01219/) (2024)
- [7th ABAW Competition: Multi-Task Learning and Compound Expression Recognition](https://mlsanity.com/paper/2407.03835/) (2024)
- [Facial Action Unit Detection Using Attention and Relation Learning](https://mlsanity.com/paper/1808.03457/) (2018)
- [Deep Structured Learning for Facial Action Unit Intensity Estimation](https://mlsanity.com/paper/1704.04481/) (2017)

## Built on by (most cited first, 1 of 1)

- [From Affect to Complex Behavior: Advancing Multimodal Human-Centered AI at the 10th ABAW Workshop & Competition](https://mlsanity.com/paper/2605.27451/) (2026)

## Abstract (arXiv)

Facial Action Unit (AU) detection in in-the-wild environments remains a formidable challenge due to severe spatial-temporal heterogeneity, unconstrained poses, and complex audio-visual dependencies. While recent multimodal approaches have made progress, they often rely on capacity-limited encoders and shallow fusion mechanisms that fail to capture fine-grained semantic shifts and ultra-long temporal contexts. To bridge this gap, we propose a novel multimodal framework driven by Hierarchical Granularity Alignment and State Space Models.Specifically, we leverage powerful foundation models, namely DINOv2 and WavLM, to extract robust and high-fidelity visual and audio representations, effectively replacing traditional feature extractors. To handle extreme facial variations, our Hierarchical Granularity Alignment module dynamically aligns global facial semantics with fine-grained local active patches. Furthermore, we overcome the receptive field limitations of conventional temporal convolutional networks by introducing a Vision-Mamba architecture. This approach enables temporal modeling with O(N) linear complexity, effectively capturing ultra-long-range dynamics without performance degradation. A novel asymmetric cross-attention mechanism is also introduced to deeply synchronize paralinguistic audio cues with subtle visual movements.Extensive experiments on the challenging Aff-Wild2 dataset demonstrate that our approach significantly outperforms existing baselines, achieving state-of-the-art performance. Notably, this framework secured top rankings in the AU Detection track of the 10th Affective Behavior Analysis in-the-wild Competition.