Machine and deep learning in facial expression recognition: a survey based on facial action units
Mauricio Priego, Itzel Aranguren, Diego Oliva, Arturo Valdivia-G, Mario A Navarro, Marco Pérez-Cisneros · 2026
Read Machine and deep learning in facial expression recognition: a survey based on facial action units by Mauricio Priego, Itzel Aranguren, Diego Oliva, Arturo Valdivia-G, Mario A Navarro, Marco Pérez-Cisneros (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13562657.txt.
Description
Mauricio Priego and five colleagues survey how machines learned to read faces between 2019 and 2024, with a special eye on the small muscle actions beneath each expression. Software that senses emotion serves security, health care, and robots that work beside people. Rather than jumping straight from a face to a feeling, many systems first detect facial action units, the coded twitches of single muscles defined by anatomy. The 2026 survey sorts the literature into four families: action-unit detection alone, direct expression reading, expression reading assisted by units, and pipelines that chain the two.
Across those families the authors compare methods, performance patterns, and the conditions where the middle step pays off. Unit-based routes shine when data is scarce or expressions are subtle, since the units give a grounded, explainable scaffold. The survey closes by mapping open problems for builders who want systems that work outside the lab.
Catalog record
| CREATOR | Mauricio Priego, Itzel Aranguren, Diego Oliva, Arturo Valdivia-G, Mario A Navarro, Marco Pérez-Cisneros |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 134233 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |