Researchers from the University of Exeter and The University of Southampton have created a machine-learning algorithm to automatically generate an accurate facial expression from a subject’s facial features.

The researchers used facial recognition software from IBM Watson to identify people’s facial expressions using facial images captured from an online facial recognition database.

The facial recognition technology uses the subject’s unique facial features, such as their eyes and nose, to create an artificial model of the person’s face.

The computer algorithm is then used to predict the facial expression of each person based on these features.

This process is then fed into the face recognition software to generate a facial expression.

The system was able to accurately identify the facial features of more than 2,000 subjects from the database of over 200,000 faces.

“The machine-to-human face recognition process has been shown to be very accurate and can be used to generate human-like facial expressions from a vast amount of data,” Professor Peter Della Valle said.

“This is an example of how a facial recognition system can learn from a wide range of data.”

The researchers say their system is still in its early stages of development but it is already able to recognise expressions that can be recognised by humans.

“Using the face database to train a machine is not difficult at all.

We can use the database as a starting point for training a neural network and then we can use a face database as an input for the network.”

It takes about ten to 20 minutes for training to complete,” said Professor Della Ville.”

We are only at the stage where we can make use of this system to generate realistic expressions.

“A good result could be a 10 to 15-minute film where a robot has been trained to produce the expression of a person.”

The research was published in the journal Computers in Human Behavior.