AI Denoising that runs in the browser.

Based on Open Image Denoise (OIDN) and powered by WebGPU + onnxruntime-web

Denoiser runs OIDN's pre-trained U-Nets fully on the GPU: ONNX models on the WebGPU execution provider, with all pre/post-processing (normalization, tiling, overlap blending, color transforms) as WGSL compute on the same GPUDevice. Images up to ~1080p denoise in a single model run — 512² in ~14 ms and 1080p in ~104 ms warm (fp16, M-series laptop) — fast enough to denoise progressively while a path tracer accumulates.

v2 note: the library was rewritten from TensorFlow.js (abandoned) to WebGPU + onnxruntime-web, with a new stateless per-call API (clean break). Coming from 0.x? See the migration guide.

Basic example

import { Denoiser } from "denoiser";

const denoiser = await Denoiser.create();
const clean = await denoiser.denoise(document.getElementById("noisy-img")); // ImageData
canvas.getContext("2d").putImageData(clean, 0, 0);

Zero-copy GPU pipeline (three.js)

Create the denoiser first (onnxruntime-web owns the GPUDevice), hand denoiser.device to your WebGPURenderer, and run render targets straight through — pathtracer → denoiser → render target, no CPU pixels:

const denoiser = await Denoiser.create({ precision: "fp16" });
const renderer = new THREE.WebGPURenderer({ device: denoiser.device });

const out = await denoiser.denoiseTextures({
  color: tracerGpuTexture,   // float, linear HDR
  albedo, normal,            // optional MRT G-buffer -> guided model auto-selected
  hdr: true,
  inputFlipY: true,          // render targets are bottom-up
  output: myStorageTexture,  // optional: resolve into a texture three.js samples
});

Docs:

• packages/denoiser/README.md — install, API sketch, device-lifetime rules.
• docs/guides/three-js-render-targets.md — render targets in/out, full parameter reference, pitfalls.
• docs/guides/migrating-from-v1.md — 0.x → 2.x mapping.
• docs/ — index: status/next actions, specs, perf results.

Examples (/examples)

• three-pathtracer-webgpu — three r185 WebGPUPathTracer → denoiser on one shared device: CPU vs zero-copy paths, live progressive denoising, MRT G-buffer aux.
• bench — the performance harness (sizes × precision × quality, PSNR parity).
• denoiser-package-test — minimal end-to-end package test.
• webgpu-ort-smoke — low-level ORT/WGSL engine harness.

Repo layout

• packages/denoiser — the library (npm: denoiser).
• tools/onnx-convert — Python tooling that builds the ONNX U-Nets directly from OIDN .tza weights (no PyTorch). Models ship fp32 + fp16 with dynamic batch/height/width dims. They are hosted separately from the library (a page only fetches the one model it uses, 0.6–15 MB) — see Hosting the models.
• docs/ — guides (three.js interop, v1→v2 migration), status/next actions, specs, and the perf plan + measured results.

Development

corepack yarn install
cd packages/denoiser && yarn build

# convert models once (serves at /models in the example dev servers)
cd tools/onnx-convert && python3 -m venv .venv && . .venv/bin/activate
pip install -r requirements.txt
python convert.py ../../packages/denoiser/tzas/*.tza -o ../../packages/denoiser/models
python convert.py ../../packages/denoiser/tzas/*.tza -o ../../packages/denoiser/models --fp16

cd examples/three-pathtracer-webgpu && yarn dev

License

MIT. OIDN weights are Apache-2.0 (© Intel) — see packages/denoiser/tzas/LICENSE.txt.