FaceReader methodology white paper

FaceReader methodology

Please enter your first name.
Please enter your last name.
Please enter a valid email address.


Please accept the Noldus privacy policy.

How does FaceReader work?

FaceReader was trained to classify the six basic or universal emotions: happy, sad, angry, surprised, scared, and disgusted. Additionally, FaceReader can recognize a ‘neutral’ state and analyze ‘contempt’. Moreover, the network was trained to classify twenty Facial Action Units. Use the custom expression tool to create your own advanced expression analyses.

  • Find: The position of the face is found using a deep learning
  • Model: FaceReader models the face based on deep neural networks
  • Classify: Expressions are classified by a trained deep artificial neural network

Common Questions

Is FaceReader based on Facial Action Coding System (FACS)?

Yes! Unlike open source tools, FaceReader has been validated against FACS-trained coders.

Does FaceReader work with any face?

Yes! FaceReader has been validated for any face and it has no biases for or against any face.

How is FaceReader validated?

FaceReader’s classification was compared with manual annotation by two certified FACS coders. FaceReader is over 97% accurate!

FaceReader methodology white paper

FaceReader methodology

Please enter your first name.
Please enter your last name.
Please enter a valid email address.


Please accept the Noldus privacy policy.