A GPU-based hybrid tree construction method in parallel

Abstract

The invention discloses a hybrid tree parallel construction method based on a GPU. A coordinate axis with the maximum patch distribution variance is selected from three coordinate axes of model space needing to be rendered, the coordinates of a location point, perpendicular to the coordinate axis, of a cutting plane are calculated, the numbers of patches on the two sides of a cut face based on the location point are made to be equal, KD tree division is carried out on scene data in the model space step by step, and then octree division is carried out on divided leaf nodes in sequence. In the process of dividing the model space, the GPU technology is used for accelerating processing, and the scene division speed is greatly improved. Due to the fact that selection division based on the KD tree is firstly carried out on three dimensions, divided scenes have the advantage that patches on the three dimensions are evenly distributed, quality guarantees are provided for the subsequent fast octree division, invalid traversal and intersection operation are greatly reduced, and the method is especially suitable for calculating visibility of nonuniform complex scenes.

Description

technical field [0001] The invention relates to the technical field of graphics real-time rendering, in particular to a GPU-based hybrid tree parallel construction method. Background technique [0002] Virtual reality technology (Virtual Reality-VR) technology, also known as spirit world technology or artificial environment, was first proposed by Professor Jaron Lanier of the United States in the 1980s. It mainly uses computer simulation to generate a three-dimensional virtual world and provides users with simulations of vision, hearing, touch and other senses. It is a technology for computers to visualize and interact with complex data. [0003] Ray tracing technology is a widely used technology in the field of graphics rendering. Its biggest advantage is to generate high-quality images and draw high-realistic smooth reflection, refraction, soft shadow and other global illumination effects, but its calculation is very high, so in the past this technology can only be applie...

AI Technical Summary

Problems solved by technology

[0006] The traditional octree structure divides the scene according to the middle position of the object in the three vertical coordinate axes x, y, and z directions. Although this division method is simple and fast, its rough quality causes a large number of invalid traversal and intersection operations , also caused a large number of empty nodes and wasted storage space, making the octree gradually replaced by the KD tree with higher construction quality

Although the division of the KD tree has efficient structural features, the computational complexity of its division is much higher than that of the octree, which makes the preprocessing time of the division difficult to meet the real-time ray tracing calculation requirements of dynamic scenes.

On the other hand, the current GPU architecture contains multiple multi-core processors, and it is necessary to run tens of thousands of threads at the same time to fully utilize the computing power of these processors. However, the generation speed of nodes in the construction process of acceleration structures such as KD trees is slow, which greatly GPU computing resources are wasted, which in turn affects the construction speed

Images