The recent innovation of graphics processing unit (GPU) improved the calculation performance to be fast enough to
realize realtime breeding of animations on the personal computer. SBArt4 is an extended version of design tool of abstract
2D image by means of Interactive Evolutionary Computation. The new version compiles each expression in genotype into a
type of shading language, Core Image kernel language, that directly runs on GPU. It achieved enough speed for users to
evaluate the products of abstract animations immediately. The compiled code can be exported to another application that
utilizes Core Image framework on MacOS X, such as video effect plug-ins for video authoring tool and an independent
application for slide presentation. It is useful not only to create an abstract animation in arbitrary size and duration
but also to make a transition effect by deformation and/or discoloration. It also has potential to breed force fields
to create an aesthetic motion of particles and swarm.
About the Speaker : Dr. Tatsuo Unemi graduated from Department of Control Engineering, Tokyo Institute of Technology in 1978,
received Master's degree from Department of System Sciences in 1980, and Doctor's degree in 1994 from the same university.
He worked as a research associate from 1981 to 1987 at Tokyo Institute of Technology, as an assistant professor at
Nagaoka University of Technology, as an assistant professor at Soka University from 1992, and then the current position from 1995.
He was also working at Laboratory for International Fuzzy Engineering Research from 1992 to 1995 as a visiting scholar. He was
staying at AI Lab., IFI, University of Zurich as a visiting professor from April to September 2000. His research included Natural
Language Processing, Knowledge Engineering, and Machine Learning. Current interests include Genetic Algorithm, Reinforcement
Learning, Distributed Autonomous Robot System, and Artificial Life.
