WebSplatter: Enabling Cross-Device Efficient Gaussian Splatting in Web Browsers via WebGPU
🤖 AI Summary
This work addresses the challenges of efficiently rendering Gaussian splatting on the web, where WebGPU’s lack of global atomic operations and inherent memory and performance bottlenecks hinder cross-device scalability. To overcome these limitations, the authors propose an end-to-end GPU rendering pipeline that innovatively replaces atomic operations with a wait-free hierarchical radix sort. Additionally, they introduce an opacity-aware dynamic geometry culling mechanism that substantially reduces overdraw and peak memory consumption. Experimental results demonstrate that the proposed method achieves 1.2× to 4.5× speedups over existing web-based Gaussian splatting renderers across a range of devices, significantly improving both rendering efficiency and memory utilization.
Application Category
📝 Abstract
We present WebSplatter, an end-to-end GPU rendering pipeline for the heterogeneous web ecosystem. Unlike naive ports, WebSplatter introduces a wait-free hierarchical radix sort that circumvents the lack of global atomics in WebGPU, ensuring deterministic execution across diverse hardware. Furthermore, we propose an opacity-aware geometry culling stage that dynamically prunes splats before rasterization, significantly reducing overdraw and peak memory footprint. Evaluation demonstrates that WebSplatter consistently achieves 1.2$\times$ to 4.5$\times$ speedups over state-of-the-art web viewers.
Problem
Research questions and friction points this paper is trying to address.
Gaussian Splatting
WebGPU
cross-device rendering
web browser
efficient rendering
Innovation
Methods, ideas, or system contributions that make the work stand out.
Gaussian Splatting
WebGPU
wait-free sorting
opacity-aware culling
cross-device rendering
🔎 Similar Papers
2024-01-08arXiv.orgCitations: 127
