Cone Based Tree to Improve Ray Tracing

Abstract

Real-time ray tracing has risen to be the topic of research in recent years. Different approaches have been introduced to optimize the algorithm. This study presented a new type of data structure, cone based i.e., CB-tree for traditional ray tracing. This data structure is applied for primary ray and shadow ray optimization. The cone’s definition has been modified by including information on the vertex, reference base point, and radius with the height of the cone. The advantages of this approach are that it avoids unnecessary ray object interaction calculation for both primary ray and shadow ray and finally, this is a procedure for faster image rendering calculation by maintaining the same level of details as traditional ray tracing. The study reveals that the proposed method CB-Tree shows the ratio of 0.021 and .007 for hard and soft shadow rays respectively, which is a much-improved performance than the traditional approach ratio 1. (performance would be good when less than 1).