# A Transfer Learning approach to Heatmap Regression for Action Unit intensity estimation

arXiv: https://arxiv.org/abs/2004.06657 · HTML: https://mlsanity.com/paper/2004.06657/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Apr 2020.

## Built on by (most cited first, 10 of 10)

- [Pre-training strategies and datasets for facial representation learning](https://mlsanity.com/paper/2103.16554/) (2021)
- [COLD Fusion: Calibrated and Ordinal Latent Distribution Fusion for Uncertainty-Aware Multimodal Emotion Recognition](https://mlsanity.com/paper/2206.05833/) (2022)
- [Affective Processes: stochastic modelling of temporal context for emotion and facial expression recognition](https://mlsanity.com/paper/2103.13372/) (2021)
- [Facial Action Unit Detection and Intensity Estimation from Self-supervised Representation](https://mlsanity.com/paper/2210.15878/) (2022)
- [FG-Net: Facial Action Unit Detection with Generalizable Pyramidal Features](https://mlsanity.com/paper/2308.12380/) (2023)
- [Are 3D Face Shapes Expressive Enough for Recognising Continuous Emotions and Action Unit Intensities?](https://mlsanity.com/paper/2207.01113/) (2022)
- [Automatic Facial Paralysis Estimation with Facial Action Units](https://mlsanity.com/paper/2203.01800/) (2022)
- [AUEditNet: Dual-Branch Facial Action Unit Intensity Manipulation with Implicit Disentanglement](https://mlsanity.com/paper/2404.05063/) (2024)
- [MAUGen: A Unified Diffusion Approach for Multi-Identity Facial Expression and AU Label Generation](https://mlsanity.com/paper/2602.00583/) (2026)
- [DIME: Scaling Facial Representation Learning via Differential Masked Autoencoding](https://mlsanity.com/paper/2610.17015/) (2026)

## Abstract (arXiv)

Action Units (AUs) are geometrically-based atomic facial muscle movements known to produce appearance changes at specific facial locations. Motivated by this observation we propose a novel AU modelling problem that consists of jointly estimating their localisation and intensity. To this end, we propose a simple yet efficient approach based on Heatmap Regression that merges both problems into a single task. A Heatmap models whether an AU occurs or not at a given spatial location. To accommodate the joint modelling of AUs intensity, we propose variable size heatmaps, with their amplitude and size varying according to the labelled intensity. Using Heatmap Regression, we can inherit from the progress recently witnessed in facial landmark localisation. Building upon the similarities between both problems, we devise a transfer learning approach where we exploit the knowledge of a network trained on large-scale facial landmark datasets. In particular, we explore different alternatives for transfer learning through a) fine-tuning, b) adaptation layers, c) attention maps, and d) reparametrisation. Our approach effectively inherits the rich facial features produced by a strong face alignment network, with minimal extra computational cost. We empirically validate that our system sets a new state-of-the-art on three popular datasets, namely BP4D, DISFA, and FERA2017.