Empathic Painting
Boston University and University of Bath
Maria Shugrina, Margrit Betke, John Collomosse

Empathic Painting is a video-based human-computer interface that allows an artist to use facial expressions to affect the output of a rendering algorithm. With the interface, the artist converts an input photograph into a painting with stroke patterns and colors that are selected based on the artist's facial expression. The interface uses a suite of Computer Vision algorithms capable of recognizing facial expressions through the detection of "facial action units." Detection results are mapped to vectors within a continuous 2D space representing emotional state. State vectors are then used to compute style parameters of a fast segmentation-based painterly rendering algorithm. The result is a digital canvas capable of smoothly varying its painterly style at approximately 4 frames per second, providing a novel user interactive experience using only commodity hardware.



Copyright, Shugrina, Betke, Collomosse, 2006.