Evolutionary optimization of functionally defined shapes: case study of natural optical objects

Abstract

The paper focuses on an approach to modeling shapes through the use of evolutionary optimization or genetic algorithms for functionally represented geometric objects. This representation allows us to take into account non-manifold geometry and material distribution inside natural lenses. We use ray tracing as an instrument of the current shape evaluation. We describe a case study of a "four-eyed" fish and discuss problems of using the optimization procedure.