One of the most important elements of a human-to-human communication analysis system is knowing what has the attention of the people (meeting participants). Using machine learning models based on the orientation of participant heads, the AMI project is developing technology that identifies who has the focus of attention (and who or what does not). In this demonstration movie we show that the model is very sensitive to the head orientation and highly reliable at predicting the participant’s focus of attention. [movie]

Click on the image to see full size