Building An Efficient Grid On GPU

TL;DR

This work presents a parallel grid construction algorithm which can efficiently construct a structured grid on GPU hardware, and is substantially faster than existing uniform grid construction algorithms, especially on non-homogeneous scenes.

Abstract

Grid space partitioning is a technique to speed up queries to graphics databases. We present a parallel grid construction algorithm which can efficiently construct a structured grid on GPU hardware. Our approach is substantially faster than existing uniform grid construction algorithms, especially on non-homogeneous scenes. Indeed, it can populate a grid in real-time (at rates over 25 Hz), for architectural scenes with 10 million triangles.

Figures (5)

• Figure 1: Bounding box cells for two dissimilar triangles with 1 and 12 cells respectively
• Figure 2: Example grid
• Figure 3: Object boxes with Cell IDs in Relative (at left in blue) or Global (at right in red) cell offsets
• Figure 4: Grid construction for example scene
• Figure 5: Build times (less is better) for selected scenes