Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Physically-based Path Tracer using WebGPU and OpenPBR

Simon Stucki, Philipp Ackermann

TL;DR

The paper presents a web-based, open-source path tracer that leverages WebGPU and the OpenPBR shading model to render photorealistic CAD scenes directly in the browser, addressing rasterization's poor global illumination on the web. It details a GPU-accelerated pipeline with CPU-side scene preparation (BVH) and a backward ray-tracing ray generator, OpenPBR-based shading, and progressive rendering with tone mapping. The results demonstrate near real-time rendering capabilities for complex architectures and configurable materials, using glTF inputs and producing visually faithful results without pregenerated assets. This approach reduces hosting costs and enables interactive, browser-based visualization of configurable industrial products, highlighting WebGPU's potential for advanced web rendering and production workflows.

Abstract

This work presents a web-based, open-source path tracer for rendering physically-based 3D scenes using WebGPU and the OpenPBR surface shading model. While rasterization has been the dominant real-time rendering technique on the web since WebGL's introduction in 2011, it struggles with global illumination. This necessitates more complex techniques, often relying on pregenerated artifacts to attain the desired level of visual fidelity. Path tracing inherently addresses these limitations but at the cost of increased rendering time. Our work focuses on industrial applications where highly customizable products are common and real-time performance is not critical. We leverage WebGPU to implement path tracing on the web, integrating the OpenPBR standard for physically-based material representation. The result is a near real-time path tracer capable of rendering high-fidelity 3D scenes directly in web browsers, eliminating the need for pregenerated assets. Our implementation demonstrates the potential of WebGPU for advanced rendering techniques and opens new possibilities for web-based 3D visualization in industrial applications.

Physically-based Path Tracer using WebGPU and OpenPBR

TL;DR

The paper presents a web-based, open-source path tracer that leverages WebGPU and the OpenPBR shading model to render photorealistic CAD scenes directly in the browser, addressing rasterization's poor global illumination on the web. It details a GPU-accelerated pipeline with CPU-side scene preparation (BVH) and a backward ray-tracing ray generator, OpenPBR-based shading, and progressive rendering with tone mapping. The results demonstrate near real-time rendering capabilities for complex architectures and configurable materials, using glTF inputs and producing visually faithful results without pregenerated assets. This approach reduces hosting costs and enables interactive, browser-based visualization of configurable industrial products, highlighting WebGPU's potential for advanced web rendering and production workflows.

Abstract

This work presents a web-based, open-source path tracer for rendering physically-based 3D scenes using WebGPU and the OpenPBR surface shading model. While rasterization has been the dominant real-time rendering technique on the web since WebGL's introduction in 2011, it struggles with global illumination. This necessitates more complex techniques, often relying on pregenerated artifacts to attain the desired level of visual fidelity. Path tracing inherently addresses these limitations but at the cost of increased rendering time. Our work focuses on industrial applications where highly customizable products are common and real-time performance is not critical. We leverage WebGPU to implement path tracing on the web, integrating the OpenPBR standard for physically-based material representation. The result is a near real-time path tracer capable of rendering high-fidelity 3D scenes directly in web browsers, eliminating the need for pregenerated assets. Our implementation demonstrates the potential of WebGPU for advanced rendering techniques and opens new possibilities for web-based 3D visualization in industrial applications.
Paper Structure (17 sections, 1 equation, 5 figures, 2 tables)

This paper contains 17 sections, 1 equation, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Ray Tracing
  3. Use Cases
  4. Prior Work
  5. Physically Based Rendering
  6. Results
  7. Implementation
  8. CPU Setup
  9. Ray Generator
  10. Path Tracer
  11. Render Pipeline
  12. Performance
  13. Discussion
  14. Real-Time Rendering
  15. Production Readiness of WebGPU
  16. ...and 2 more sections

Figures (5)

  • Figure 1: A two-step pre-processing stage is employed for offline as well as real-time rendering pipelines.
  • Figure 2: In an offline rendering setup, the number of images grows exponentially. The number of images increases further when offering 360° viewing.
  • Figure 3: A real-time rendering pipeline solely relies on having an adequate model for each component of the assembly. It does not require pre-rendering every possible configuration.
  • Figure 4: Path Tracer Pipeline, distinguishing GPU and CPU tasks. The stages of the compute pipeline and render pipeline are executed sequentially.
  • Figure 5: The three LOD artifacts, from left to right: 1,068,735 triangles, 106,873 triangles, 10,687 triangles. The left and middle figures share similar visual fidelity characteristics.