How to set up FaceReader

FaceReader analyzes facial expressions from a camera or from images/video files you already have. You can run it in your own lab, collect data online, or build it into your software. Whichever setup you choose, a camera and a computer are enough to get started.

On this page you’ll find:

  • checkThe setup options and the hardware you need for each option
  • checkThe models and license options
  • checkInformation on privacy and ethics in FaceReader
Contact us arrow_downward
Social Media Lab Mons, Belgium

What computer hardware do I need?

Every setup requires a camera and a computer. For more details, download the technical specifications. The table below shows what each option requires.

Stationary lab FaceReader Online SDK integration
Camera Fixed camera or webcam Participant’s own webcam Camera or video source
Computer Windows computer for analysis Windows computer for analysis Windows or Android device, PC or server
Optional Laptop or screen, Lightning equipment – –

Minimum requirements: Windows 11 Pro (64-bit), Intel Core i5 (8th Gen) / AMD Ryzen 5 (4000 series), 8 GB RAM, and AVX2 support.

Choose your model

Faces from baby to adult across various ethnicities

FaceReader offers models for different participant groups.

General – The standard model for most studies and participants.

East Asian – Optimized for East Asian faces, for more accurate analysis of participants of East Asian descent.

Baby – For infants between 6 and 24 months old. Based on the Baby Facial Action Coding System (Baby FACS) and includes 14 Action Units specific to infants.

Get the best results from your recordings

Good recordings lead to more reliable results. Read how to optimize your analysis

Camera at eye level

Place the camera at eye level or just below the screen.

Face the participant

Point the camera straight at the participant’s face.

Light from the front

Light the face evenly from the front.

Avoid backlight

No bright light behind the face, no strong shadows.

Privacy & ethics in FaceReader

Signing an informed consent form on a tablet

FaceReader is installed on-site and follows strict privacy-by-design principles. The software analyzes facial expressions and Action Units, but it cannot recognize or identify faces or people. This makes it unsuitable for surveillance, and it cannot be used as a lie detector.

You require informed consent from every person whose facial expressions are analyzed. As the user, you decide how to obtain consent, for example with an explicit opt-in, and you make clear how the data will be used and stored.

FaceReader is intended for research purposes. Read more about our policy on ethics in facial expression analysis.

Purdue University Logo Purdue University
Harvard University Logo Harvard University
TNO Logo TNO
University of Washington Logo University of Washington
Danish Technological Institute Logo Danish Technological Institute
Universität Leipzig Logo Universität Leipzig
B-com Logo B-com
WUR Logo WUR
LSE Logo LSE
Purdue University Logo Purdue University
Harvard University Logo Harvard University
TNO Logo TNO
University of Washington Logo University of Washington
Danish Technological Institute Logo Danish Technological Institute
Universität Leipzig Logo Universität Leipzig
B-com Logo B-com
WUR Logo WUR
LSE Logo LSE

Want more info?

Curious about this product but not ready to decide yet? Just let us know what you need: more information, a price quote, a free trial, a demo, or an online meeting with one of our specialists. We're here to help you with the next steps to finding your solution.

check_circle

Thank you!

We will get back to you within one business day.

error

error

error

error

By clicking Submit, you consent to Noldus processing your data as described in our privacy policy.

shopping_bag
check_circle

Thank you!

We'll get back to you shortly.

error

Please correct the following errors:

error

error

error

error

By clicking Submit, you consent to Noldus processing your data as described in our privacy policy.

check_circle

Thank you for registering!

We look forward to seeing you at the NoldusViso launch event.

error

Please correct the following errors:

error

error

error

error

error

By clicking Register, you consent to Noldus processing your data as described in our privacy policy.