# Facial Action Unit Intensity Estimation via Semantic Correspondence Learning with Dynamic Graph Convolution

arXiv: https://arxiv.org/abs/2004.09681 · HTML: https://mlsanity.com/paper/2004.09681/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Apr 2020.

## Built on by (most cited first, 22 of 22)

- [CelebV-HQ: A Large-Scale Video Facial Attributes Dataset](https://mlsanity.com/paper/2207.12393/) (2022)
- [Facial Action Unit Detection and Intensity Estimation from Self-supervised Representation](https://mlsanity.com/paper/2210.15878/) (2022)
- [Graph-based Facial Affect Analysis: A Review](https://mlsanity.com/paper/2103.15599/) (2021)
- [BAH Dataset for Ambivalence/Hesitancy Recognition in Videos for Digital Behavioural Change](https://mlsanity.com/paper/2505.19328/) (2025)
- [FG-Net: Facial Action Unit Detection with Generalizable Pyramidal Features](https://mlsanity.com/paper/2308.12380/) (2023)
- [Guided Interpretable Facial Expression Recognition via Spatial Action Unit Cues](https://mlsanity.com/paper/2402.00281/) (2024)
- [Two-stage Temporal Modelling Framework for Video-based Depression Recognition using Graph Representation](https://mlsanity.com/paper/2111.15266/) (2021)
- [Multi-Task Learning Framework for Emotion Recognition in-the-wild](https://mlsanity.com/paper/2207.09373/) (2022)
- [Graph Neural Networks in Computer Vision -- Architectures, Datasets and Common Approaches](https://mlsanity.com/paper/2212.10207/) (2022)
- [Feature Pyramid Network for Multi-task Affective Analysis](https://mlsanity.com/paper/2107.03670/) (2021)
- [Multimodal Ambivalence/Hesitancy Recognition in Videos for Personalized Digital Health Interventions](https://mlsanity.com/paper/2604.11730/) (2026)
- [AuE-IPA: An AU Engagement Based Infant Pain Assessment Method](https://mlsanity.com/paper/2212.04764/) (2022)
- [One-Frame Calibration with Siamese Network in Facial Action Unit Recognition](https://mlsanity.com/paper/2409.00240/) (2024)
- [EventFormer: AU Event Transformer for Facial Action Unit Event Detection](https://mlsanity.com/paper/2203.06355/) (2022)
- [Micro-Expression Recognition Based on Attribute Information Embedding and Cross-modal Contrastive Learning](https://mlsanity.com/paper/2205.14643/) (2022)
- [Automated Deception Detection from Videos: Using End-to-End Learning Based High-Level Features and Classification Approaches](https://mlsanity.com/paper/2307.06625/) (2023)
- [Occlusion Aware Student Emotion Recognition based on Facial Action Unit Detection](https://mlsanity.com/paper/2307.09465/) (2023)
- [Leveraging Previous Facial Action Units Knowledge for Emotion Recognition on Faces](https://mlsanity.com/paper/2311.11980/) (2023)
- [FERGI: Automatic Scoring of User Preferences for Text-to-Image Generation from Spontaneous Facial Expression Reaction](https://mlsanity.com/paper/2312.03187/) (2023)
- [MAUGen: A Unified Diffusion Approach for Multi-Identity Facial Expression and AU Label Generation](https://mlsanity.com/paper/2602.00583/) (2026)
- [Facial-Expression-Aware Prompting for Empathetic LLM Tutoring](https://mlsanity.com/paper/2604.15336/) (2026)
- [Capturing Dynamics: The 4D Facial Expression Intensity Dataset](https://mlsanity.com/paper/2610.02647/) (2026)

## Abstract (arXiv)

The intensity estimation of facial action units (AUs) is challenging due to subtle changes in the person's facial appearance. Previous approaches mainly rely on probabilistic models or predefined rules for modeling co-occurrence relationships among AUs, leading to limited generalization. In contrast, we present a new learning framework that automatically learns the latent relationships of AUs via establishing semantic correspondences between feature maps. In the heatmap regression-based network, feature maps preserve rich semantic information associated with AU intensities and locations. Moreover, the AU co-occurring pattern can be reflected by activating a set of feature channels, where each channel encodes a specific visual pattern of AU. This motivates us to model the correlation among feature channels, which implicitly represents the co-occurrence relationship of AU intensity levels. Specifically, we introduce a semantic correspondence convolution (SCC) module to dynamically compute the correspondences from deep and low resolution feature maps, and thus enhancing the discriminability of features. The experimental results demonstrate the effectiveness and the superior performance of our method on two benchmark datasets.