To gain accurate and reliable data about facial expressions, FaceReader is the most robust automated system that will help you out.
Clear insights into the effect of different stimuli on emotions
Very easy to use: save valuable time and resources
Easy integration with eye tracking data and physiology data
Trusted by researchers around the world
FaceReader benefits your work
Many researchers have turned towards using automated facial expression analysis software to better provide an objective assessment of emotions. FaceReader™ is used worldwide at more than 1,000 universities, research institutes, and companies in many markets, from consumer and psychology research to usability studies.
FaceReader software is fast, flexible, objective, accurate, and easy to use. It immediately analyzes your data (live, video, or still images), saving valuable time. The option to record audio as well as video makes it possible to hear what people have been saying – for example, during human-computer interactions, or while watching stimuli.
Measure emotions with FaceReader
FaceReader is the best automated system for the recognition of specific properties in facial images and expressions. Aside from the basic or universal facial expressions, you can define your own Custom Expressions. Additionally, FaceReader can recognize a neutral state and analyze contempt.
Whether your test participant is a baby, child, adult, or older person, FaceReader adjusts the analysis to the model that best fits your research.
How FaceReader works
- Face finding – Finds a face using a face-finding algorithm based on Deep Learning
- Face modeling – Makes an accurate artificial face model using almost 500 key points
- Face classification – Classifies facial expressions with artificial neural networks
As a result, you'll have data on basic and custom expressions, head orientation, gaze direction, valence and arousal, Action Units, heart rate and heart rate variability, and consumption behavior.
Accurate emotion classification
According to a validation study using the ADFES data set, FaceReader 9 delivers accurate performance for emotion classification, with an average accuracy of 99% .