Eye tracking is a technique where a person’s eye movements, defined as the gaze, is recorded via a camera. This movement can be captured either remotely like when a person is looking at a computer-mounted camera, or else by wearing a head-mounted camera device. One of the main uses of eye tracking is to understand where a person’s gaze falls when looking at an object.
As an example, let us consider a standard display in a shop. As individuals, we tend to first look at the overall exhibit and then focus on one particular object or part of an object. There are times when we are conscious that we are focusing on a particular object that we like but there are others when whatever holds our attention is so subtle that we don’t even realise that we are looking at it more in-depth. But the application of eye tracking is not just for shop displays or for marketing purposes.
Think about the ways that you are using your eyes right now, and you will realise that in just a few minutes you have used your sight to read this article, possibly typing in words on a computer to seeing what you have been writing, blinked, got distracted by something else that caught your attention and your gaze... The possibilities of capturing eye movement data are endless, from the medical to the entertainment and educational applications, to mention but a few.
At the University of Malta, the Department of Artificial Intelligence and the Department of Digital Arts, have merged efforts to track eye movement on works of art. What kicked off the concept for this project was the artist’s question about how people perceive his work of art involving line sculptures. We were asking questions like: What do people focus on when viewing line sculptures? How do they see the art taking form? Can we understand their perception of the art through the movements of their eyes and the points where they focus the most? Eye-tracking art is not without its challenges however.
Drawing with the eyes becomes challenging because of the various eye movements that may affect the way the object is drawn. Because the art that people perceive needs to be as true to the form as to that which they are seeing, technology needs to find a way to present the form of art in a way which conveys meaning to the artist, as an interpretation of what people are seeing in the art form.
The project ‘Eye-Tracking Art’ has already shown promise of being able to overcome this challenge but it is also indicating that with more research work in the area, we can provide an alternative means of making art more accessible to individuals who may want to experience art from a different perspective.
The work presented in this article is being carried out by Dr Vanessa Camilleri (AI), Dr Vince Briffa (DA), Matthew Attard (DA), Neil Mizzi (AI) and Dr George Azzopardi (AI) from the departments of Artificial Intelligence (AI) and Digital Arts (DA) at the University of Malta.
Did you know?
• Humans and dogs are the only species known to seek visual cues from another individual’s eyes, and dogs only do this when interacting with humans.
• ‘Red eye’ occurs in photos because light from the flash bounces off the back of the eye. The choroid is located behind the retina and is rich in blood vessels, which make it appear red on film.
• An eye cannot be transplanted. More than one million nerve fibres connect each eye to the brain and currently it is not possible to reconstruct those connections.
• Your tears have different compositions based on whether something’s irritating your eye, or you’re crying, or yawning.
• A fingerprint has 40 unique characteristics, but an iris has 256, a reason retina scans are increasingly being used for security purposes.
For more trivia see: www.um.edu.mt/think
Sound bites
• Artificial Intelligence is making headway in allowing adaptable user interfaces for mobile keyboards. The idea is to model the needs of various disabilities, e.g. tremor, sight problems, or even dementia. The model uses techniques to predicthow individual abilities affect text entry on a touchscreen. The model combines psychological research on finger pointing and eye movements, allowing it to predict text entry speed, typing errors, and proofreading.
https://www.sciencedaily.com/releases/2018/03/180327093925.htm
• Engineers have developed a non-invasive approach to permanently correct vision that shows great promise in preclinical models. The method uses a femtosecond oscillator for selective and localised alteration of the biochemical and biomechanical properties of corneal tissue. The technique, which changes the tissue’s macroscopic geometry, is non-surgical and has fewer side effects and limitations than those seen in refractive surgeries. The study could lead to treatment for myopia, hyperopia, astigmatism, and irregular astigmatism.
https://www.sciencedaily.com/releases/2018/05/180529132012.htm
• For more interesting science news listen to Radio Mocha every Saturday at 11.05am on Radju Malta 93.7FM.
https://www.facebook.com/RadioMochaMalta/