FaceReader 10 - Introduction to Action Unit Classification
Last updated: Jul 28, 2026
The Action Unit Module
Main Topics
- Introduction to Action Unit classification
- Analyze Action Units
- Visualize Action Units
- Action Units in the log files
- Custom Expressions
Introduction to Action Unit Classification
Action Units are muscles or muscle groups in the face that are responsible for facial expressions. The Action Units are described in the Facial Action Coding System (FACS) published by Ekman and Friesen (1978).
FaceReader can analyze Action Units with the Action Unit Module. The following Action Units are analyzed:
| Action Unit | FACS Name |
|---|---|
| 1 | Inner Brow Raiser |
| 2 | Outer Brow Raiser |
| 4 | Brow Lowerer |
| 5 | Upper Lid Raiser |
| 6 | Cheek Raiser |
| 7 | Lid Tightener |
| 9 | Nose Wrinkler |
| 10 | Upper Lid Raiser |
| 12 | Lip Corner Puller |
| 14 | Dimpler |
| 15 | Lip Corner Depressor |
| 17 | Chin Raiser |
| 18 | Lip Pucker |
| 20 | Lip Stretcher |
| 23 | Lip Tightener |
| 24 | Lip Pressor |
| 25 | Lips Part |
| 26 | Jaw Drop |
| 27 | Mouth Stretch |
| 43 | Eyes Closed |
If you have Baby FaceReader, you can analyze the combined activation of Action Unit 3 and 4 (Brow Knitting and Knotting).
Intensities are annotated by appending letters:
- A - trace.
- B - slight.
- C - pronounced.
- D - severe.
- E - max.
These intensities follow the classification as described in Ekman et al. (2002).
In FaceReader 9.1.8 and higher the classification of Action Units gives slightly lower intensities than in older versions of the software. In older versions non-active Action Units could be classified with intensity level A, A-level Action Unit activations with level B, etc. The reason for this difference is an error in the conversion script of the underlying deep neural network that has been corrected in FaceReader 9.1.8.
The values of Contempt, Arousal, facial states (except Gaze direction) and Valence in Baby FaceReader are also affected as these parameters are based on Action Units. Valence in the regular version of FaceReader is based on facial expressions only and is, therefore, not affected. The calculation of most custom expressions is also based on Action Units and the intensities of these custom expressions are, thus, also different. For the most accurate Action Unit estimation it is important to use the latest version of the software and re-analyze old videos if you want to compare the data with those obtained with the new FaceReader version.
References
- Ekman P. and W. V. Friesen (1978). Facial Action Coding System: A Technique for the Measurement of Facial Movement. Consulting Psychologists Press, Palo Alto.
- Ekman P., W. V. Friesen, and J. C. Hager (2002). The Facial Action Coding System. Second edition. Salt Lake City: Research Nexus eBook. London: Weidenfeld & Nicolson (world).
Source: FaceReader 10 Reference Manual (Help), Noldus Information Technology