upscalejs

Image upscaling using super resolution AI models.

Usage

import { Upscaler } from "upscalejs"

const upscaler = new Upscaler()
const result = await upscaler.upscale(bitmap)
const canvas = document.createElement("canvas")
canvas.width = result.width
canvas.height = result.height
canvas.getContext("2d")?.putImageData(result, 0, 0)

Node

const fs = require("fs")
const { Upscaler, canvas } = require("upscalejs")

const upscaler = new Upscaler({
  base: "./node_modules/upscalejs/dist/"
})

const upscale = async (input, output) => {
  const img = await canvas.loadImage(input)
  const result = await upscaler.upscale(img)
  const write_stream = fs.createWriteStream(output)
  const upscale_canvas = canvas.createCanvas(result.width, result.height)
  upscale_canvas.getContext("2d").putImageData(result, 0, 0)
  const jpeg_stream = upscale_canvas.createJPEGStream()
  jpeg_stream.pipe(write_stream)
}

upscale("image.jpg", "upscaled.jpg")

You will need to copy the models and onnxruntime wasm files to your public folder e.g.:

const CopyPlugin = require("copy-webpack-plugin");

module.exports = {
  chainWebpack: config => {
    config
      .plugin("copy-webpack-plugin")
      .use(CopyPlugin)
      .tap(() => {
        return [{
          patterns: [
            {
              from: "./node_modules/upscalejs/dist/js/ort-*.wasm",
              to: "js/[name][ext]",
            },
            {
              from: "./node_modules/upscalejs/dist/models/*.onnx",
              to: "models/[name][ext]",
            }
          ],
        }]
      })
   }
}

Current available models were trained in anime images and are based on work done in the Real-CUGAN project.

Project Setup

npm install

Compile and Hot-Reload for Development

npm run dev

Type-Check, Compile and Minify for Production

npm run build

Lint with ESLint

npm run lint

Test with Vitest

npm run test