Rendering Point Clouds with Compute Shaders and Vertex Order Optimization

TL;DR

This paper introduces compute shader-based point cloud rendering methods that outperform traditional hardware pipelines, including a novel vertex order optimization that significantly boosts rendering efficiency across platforms.

Contribution

The paper presents new compute-based rendering approaches for point clouds and a vertex order optimization technique that enhances rendering performance and reduces viewpoint sensitivity.

Findings

Achieved rendering of 796 million points at 62-64 fps on RTX 3090.

Developed a high-quality aliasing reduction variant.

Vertex order optimization boosts GL_POINTS efficiency by up to 5x.

Abstract

While commodity GPUs provide a continuously growing range of features and sophisticated methods for accelerating compute jobs, many state-of-the-art solutions for point cloud rendering still rely on the provided point primitives (GL_POINTS, POINTLIST, ...) of graphics APIs for image synthesis. In this paper, we present several compute-based point cloud rendering approaches that outperform the hardware pipeline by up to an order of magnitude and achieve significantly better frame times than previous compute-based methods. Beyond basic closest-point rendering, we also introduce a fast, high-quality variant to reduce aliasing. We present and evaluate several variants of our proposed methods with different flavors of optimization, in order to ensure their applicability and achieve optimal performance on a range of platforms and architectures with varying support for novel GPU hardware features. During our experiments, the observed peak performance was reached rendering 796 million points (12.7GB) at rates of 62 to 64 frames per second (50 billion points per second, 802GB/s) on an RTX 3090 without the use of level-of-detail structures. We further introduce an optimized vertex order for point clouds to boost the efficiency of GL_POINTS by a factor of 5x in cases where hardware rendering is compulsory. We compare different orderings and show that Morton sorted buffers are faster for some viewpoints, while shuffled vertex buffers are faster in others. In contrast, combining both approaches by first sorting according to Morton-code and shuffling the resulting sequence in batches of 128 points leads to a vertex buffer layout with high rendering performance and low sensitivity to viewpoint changes.