Directed line classification

Abstract

It is one of the greatest challenges to improve rendering efficiency. An important issue in rendering is to develop a data structure which provides efficient visibility operations, because the most of the rendering time is spent for performing them. In this thesis, we present a new ray classification scheme, called directed line classification, to efficiently solve visibility problems. Our goal is two-fold: efficiency and generality, that is, generalizing the original ray classification scheme for wider applicability without losing its efficiency. Our scheme considerably reduces memory consumption while preserving its inherent time efficiency. Our key idea is due to the fact that the rays lying on the same line are duplicated over many cells in the ray classification scheme of Arvo and Kirk [3]. We are able to lower the dimensions of the ray space by classifying lines instead of rays. Our scheme gives much simpler-shaped, compact ray cells, that eventually accelerate visibility operations. We describe how to extend our scheme to perform various types of visibility operations and explore applications to advanced rendering such as ray tracing, radiosity, and beam tracing. We also develop a parallel ray-tracing algorithm based on our scheme for further speed up of ray tracing. Our experimental results indicate that the directed line classification greatly improve the efficiency of the rendering applications.