A small javascript library to develop WebAR apps. It is based on jsartoolkitNFT a lighter version of jsartoolkit5 only with NFT markerless technology. It uses ARnft-threejs for the rendering part.
1οΈβ£ Β Clone the repository:
git clone https://github.com/webarkit/ARnft.git
2οΈβ£ Β Install the npm packages with yarn:
yarn install
or with npm:
npm install
3οΈβ£ Β Run the node server:
npx http-server
4οΈβ£ Β Go to the examples:
http://localhost:8000/examples/arNFT_example.html
5οΈβ£ Β Point your device π± to the pinball image π a red cube will appear !
The ARnft library now includes support for SIMD (Single Instruction, Multiple Data) to enhance performance by parallelizing data processing tasks. This feature is particularly useful for applications requiring high computational power, such as augmented reality.
To see the SIMD feature in action, you can try the arNFT_simd_example.html example:
http://localhost:8000/examples/arNFT_example.simd.html
Download the zipped dist lib package from the releases page: webarkit/ARnft/releases and import it as a module:
<script type="importmap">
{
"imports": {
"three": "./js/third_party/three.js/three.module.min.js",
"arnft-threejs": "./js/ARnftThreejs.module.js",
"arnft": "./../dist/ARnft.mjs"
}
}
</script>
<script type="module">
import * as THREE from "three";
import arnft from "arnft";
const { ARnft } = arnft;
import ARnftThreejs from "arnft-threejs";
const { SceneRendererTJS, NFTaddTJS } = ARnftThreejs;
// Follow code for rendering ect. see the examples.or you can use raw.githack services (for development):
<script type="importmap">
{
"imports": {
"three": "https://cdn.jsdelivr.net/npm/three@<version>/build/three.module.min.js",
"arnft-threejs": "https://raw.githack.com/webarkit/ARnft-threejs/master/dist/ARnftThreejs.module.js",
"arnft": "https://raw.githack.com/webarkit/ARnft/master/dist/ARnft.mjs"
}
}
</script>
<script type="module">
// as the above code snippetor raw.cdn (for production, you need to add the hash):
// As the examples above import three.js, Arnft-threejs and Arnft in an importmap
"arnft": "https://rawcdn.githack.com/webarkit/ARnft/<hash>/dist/ARnft.js"
or if you want to import as a module with npm:
// In your package.json:
"devDependencies": {
"@webarkit/ar-nft": "^0.14.11"
},
// Then in your .ts or .js file
import { ARnft } from "@webarkit/ar-nft";Test the examples in the /examples folder:
arNFT_autoupdate_example.htmlExample with the autoupdate routine.arNFT_container_example.htmlExample with an alternative container.arNFT_event_example.htmlExample with objVisibility and eventListener.arNFT_example.htmlThe simplest example displaying a red cube.arNFT_simd_example.htmlExample with SIMD feature.arNFT_gltf_brave_robot_example.htmlMore advanced example with a gltf model and threejs events.arNFT_gltf_example.htmlExample showing a gltf model (Duck).arNFT_gltf_flamingo_example.htmlExample showing an animated gltf model (Flamingo).arNFT_image_example.htmlExample showing an image.arNFT_initialize_raw_example.htmlExample using the custom initialize function for the CameraRenderer (video).arNFT_multi_example.htmlExample with multi NFT markers.arNFT_multi_dispose_example.htmlExample with multi NFT markers and disposing worker.arNFT_multi_one_worker_example.htmlExample with multi NFT markers in one Worker.arNFT_video_example.htmlExample showing a video.arNFT_zft_example.htmlExample showing a simple cube, loading a .zft file.
You can try also a live example with React at this link: kalwalt.github.io/ARnft-ES6-react/
You can build the docs with this command:
yarn docs
Then run a live server and go to the docs' folder.
- NFT (Natural Feature Tracking) Markers, read my article: NFT natural feature tracking with jsartoolkit5
- ZFT compressed NFT markers with .zft extension, with faster loading time.
- SIMD (Single Instruction, Multiple Data) support for enhanced performance.
- ES6 standard. You can install it as a npm package and use it as a module (experimental). Install it with npm:
npm i @webarkit/ar-nft
or with yarn:
yarn add @webarkit/ar-nft
-
Configuration data in an external .json file.
-
Filtering of the matrix with the OneEuroFilter.
We are using Prettier as code formatter. You only need to run yarn format to write the formatted code with Prettier. If you want to check if the code is well formatted run instead: yarn format-check
If you make changes to the code, run these commands to build the distribution library, install all the dependencies with:
yarn --include=dev i
For a development build, that is the code will be rebuilt for every change, run:
yarn dev-ts
Instead for a production build, with more optimizations in the code and smaller size, run:
yarn build-ts

