In our culture, death is a taboo subject that we carefully avoid talking about. Yet we all know that we will one day be gone, as do our loved ones and loved ones. Except when a doctor indicates that a person has only so much life expectancy left, we never know when death will affect a loved one. Death Mask is precisely an application developed in augmented reality which gives, in real time, an estimate of a person's life expectancy.
As you can see in the video below, Death Mask augmented reality app places a clock showing a person's life expectancy after scanning their face. The experience is based on death masks, sculptures taken from the faces of people who in some cultures are believed to symbolize death. Concretely, the application will, first of all, scan a person's face in real time and then generate a virtual mask of it.
Once this operation has been carried out, Death Mask will analyze, thanks to the AgeNet machine learning model, the calculated virtual mask. The AgeNet model divides the facial image into smaller pieces and applies different filters to identify the contours of the face. This process allows not only to predict the age of the person but by comparing with his mask of death to predict the age at which the person will die.
Obviously, the estimate of the life expectancy of the person generated by the Death Mask application is far from precise. According to Fleisher, one of its authors: "". The developer adds: "".
The objective is therefore to promote dialogue and discussion on this particularly taboo subject which is death, but also on the use of Deep Learning.. The application was developed with Apple's ARKit and is not available to the general public. It was above all an experiment. There is no doubt, however, that such applications, much more precise, could one day see the light of day, in particular with insurance companies.