Boffins at the Computer Science department of Bath University have developed a software program that paints an image adapted to the mood of individual views.
'Empathic painting', as it is known, is an interactive painterly rendering whose appearance adapts in real time to reflect the perceived emotional state of the viewer.
The researchers used a suite of computer vision algorithms capable of recognising a user’s facial expressions via a webcam. Action units are then mapped to vectors within a continuous 2D space representing emotional state, from which the researchers in turn derived a continuous mapping to the style parameters of a simple but fast segmentation-based painterly rendering algorithm.
“The result is a digital canvas capable of smoothly varying its painterly style at approximately four frames per second, providing a novel user interactive experience using only commodity hardware,” said researcher John Collomosse.
Apparently, the empathic painting research is an experiment into the feasibility of using high level control parameters – namely, emotional state – to replace the plethora of low-level constraints users must typically set to affect the output of artistic rendering algorithms.
'Sunlit wet sidewalk' provides evidence of methane rainfall on the north pole of Saturn's moon Titan
Methane rainfall indicates the start of the summer season in Titan's northern hemisphere
Scientists believe there could be other hydrides or superhydrides with super conducting properties
Resetting the telemetry circuits and associated boards brought the instrument back to operations mode
Fortnite news and updates: Flaw in Fortnite authentication could have helped attackers steal player login credentials
Attackers could have used Fortnite security flaw to buy in-game currency on players' stored credit cards