Human Interactive EEG-Based Evolutionary Image Animation

Aneta Neumann, Frank Neumann

Abstract: Evolutionary algorithms are adaptive algorithms that can alter their behaviour under changing circumstances. This makes them well-suited to act in interactive environments. We introduce a framework for human interactive animations based on EEG signals. EEG signals received through a brain computer interface headset are used as the fitness function and influence the animation based on the signals received. For our experimental investigation, we consider the recently introduced quasi-random animation process and alter the process through mutation operators that depend on the EEG signals received. Furthermore, the EEG signals trigger mutations and therefore alterations to the image animation process if their values are not in a predefined range. Our experimental study shows that a large variety of animations and intermediate images are obtainable in this way.