A Framework for Learning and Rapid Implementation of Real-Time Global Illumination Methods

Abstract

Photorealism has become a growing requirement for real-time graphics applications. Among the components needed to obtain photorealism, one of the most important is the simulation of the light transport, the result of which is called global illumination. With the growing interest in the research in this field, the need to develop a learning approach for the methods that produce real-time global illumination and a technological basis have become important topics. In this paper, we present a framework for the rapid development and testing of such methods. The framework itself does not contain their implementations, but contains a set of components that underlie them. It has been developed with a multi-pass architecture that allows the reuse of pass implementations in several technique pipelines. Moreover, this framework can be used in the process of learning real-time global illumination methods thanks to an editor with graphical user interface (GUI) that offers a rich amount of information and a set of such methods, created in the framework for demonstration purposes. We evaluated the proposed framework in two ways. First, through a comparative analysis with other frameworks, which have been created or at least can be used for educational purposes and the rapid implementation of real-time global illumination techniques. The second evaluation was made for the applicability of the framework in learning by using it as software support in the process of testing global illumination knowledge on eight participants with background in computer graphics.