# A Deep Learning Perspective on the Origin of Facial Expressions

arXiv: https://arxiv.org/abs/1705.01842 · HTML: https://mlsanity.com/paper/1705.01842/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 May 2017.

## Built on by (most cited first, 6 of 6)

- [Transformer-based Multimodal Information Fusion for Facial Expression Analysis](https://mlsanity.com/paper/2203.12367/) (2022)
- [From the Lab to the Wild: Affect Modeling via Privileged Information](https://mlsanity.com/paper/2305.10919/) (2023)
- [Using Self-Supervised Auxiliary Tasks to Improve Fine-Grained Facial Representation](https://mlsanity.com/paper/2105.06421/) (2021)
- [Milmer: a Framework for Multiple Instance Learning based Multimodal Emotion Recognition](https://mlsanity.com/paper/2502.00547/) (2025)
- [Expression Recognition Analysis in the Wild](https://mlsanity.com/paper/2101.09231/) (2021)
- [Towards Privacy-Preserving Affect Recognition: A Two-Level Deep Learning Architecture](https://mlsanity.com/paper/2111.07344/) (2021)

## Abstract (arXiv)

Facial expressions play a significant role in human communication and behavior. Psychologists have long studied the relationship between facial expressions and emotions. Paul Ekman et al., devised the Facial Action Coding System (FACS) to taxonomize human facial expressions and model their behavior. The ability to recognize facial expressions automatically, enables novel applications in fields like human-computer interaction, social gaming, and psychological research. There has been a tremendously active research in this field, with several recent papers utilizing convolutional neural networks (CNN) for feature extraction and inference. In this paper, we employ CNN understanding methods to study the relation between the features these computational networks are using, the FACS and Action Units (AU). We verify our findings on the Extended Cohn-Kanade (CK+), NovaEmotions and FER2013 datasets. We apply these models to various tasks and tests using transfer learning, including cross-dataset validation and cross-task performance. Finally, we exploit the nature of the FER based CNN models for the detection of micro-expressions and achieve state-of-the-art accuracy using a simple long-short-term-memory (LSTM) recurrent neural network (RNN).