JavaScript/WebGL glasses virtual try on widget. Real time webcam experience, robust to all lighting conditions, high end 3D PBR rendering, easy to integrate, fallback to server-side rendering
Switch branches/tags
Nothing to show
Clone or download

README.md

Glasses Virtual Try-on

With this JavaScript widget, you can offer glasses virtual try on to your users. It can be included either on a website or in a mobile web application. It is very easy to integrate: only few lines of HTML code are required. If the user has a webcam, the experience is in real-time: the user see his face like in a mirror, but with glasses on. If he refuses to share his camera or if his configuration is not compatible, nevermind, the fallback version is displayed. He can still upload a picture and it is processed server-side to add the glasses on.

Features

These are the main features of this widget :

  • real-time webcam glasses virtual try on,
  • fallback to picture based virtual try on,
  • light reconstruction (ambient + directionnal),
  • very robust to all lighting conditions,
  • mobile friendly,
  • high end 3D engine with :
    • physically based rendering (PBR),
    • raytraced shadows,
    • deferred shading,
    • temporal antialiasing.

Demonstration

You can test it with these demos :

Here is a video screenshot of the Jeeliz Sunglasses application :

About the technology

Documentation

This API uses Jeeliz WebGL Deep Learning technology to detect and track the user's face using a neural network. You can find the technical documentation in doc.pdf.

Compatibility

  • If WebGL2 is available, it uses WebGL2 and no specific extension is required,
  • If WebGL2 is not available but WebGL1, we require either OES_TEXTURE_FLOAT extension or OES_TEXTURE_HALF_FLOAT extension,
  • If WebGL2 is not available, and if WebGL1 is not available or neither OES_TEXTURE_FLOAT or OES_HALF_TEXTURE_FLOAT are implemented, the user is not compatible with the real time video version.

In all cases, WebRTC should be implemented in the web browser, otherwise FaceFilter API will not be able to get the webcam video feed. Here are the compatibility tables from caniuse.com here: WebGL1, WebGL2, WebRTC.

If a compatibility error is triggered, please post an issue on this repository. If this is a problem with the webcam access, please first retry after closing all applications which could use your device (Skype, Messenger, other browser tabs and windows, ...). Please include :

If the user cannot use the realtime video version because its implementation of WebGL is too minimalistic or if he does not want to share his camera, the fallback version is displayed. On the fallback version the user upload a picture (or take a picture with his camera if he uses a mobile device) and the picture is processed server-side to overlay the glasses.

Hosting

HTTPS only !

This widget requires the user's webcam video stream through MediaStream API. So your application should be hosted by a HTTPS server (even with a self-signed certificate). It won't work at all with unsecure HTTP, even locally with some web browsers.

The scripts

You can use our hosted and up to date version of the library, available here :

https://appstatic.jeeliz.com/jeewidget/JeelizNNCwidget.js

which fetches the neuron network hosted here :

https://appstatic.jeeliz.com/jeefit/built/jeefitNNC.json

They are hosted in a content delivery network (CDN) using gzip compression. The script file is the same than release/JeelizNNCwidget.js.

Glasses 3D models and materials

Glasses models and materials are stored on the Jeeliz GlassesDB. Each model is designed by a unique SKU. You can check the different models available by opening glassesSKU.csv file. If you want us to add specific glasses models, please contact us at contact__at__jeeliz.com or here.

So this library is not fully client side because the assets are hosted by our servers. If you want to manage your own assets you can use our generalistic face detection and tracking API, Jeeliz FaceFilter API. There is a demonstration included in that repository, using THREE.JS as 3D rendering engine, which can be a starting point, Miel Pops Glasses.

License

Apache 2.0. This application is free for both commercial and non-commercial use.

We appreciate attribution by including the Jeeliz logo and a link to the Jeeliz website in your application or desktop website. Of course we do not expect a large link to Jeeliz over your face filter, but if you can put the link in the credits/about/help/footer section it would be great.

See also

Jeeliz main face detection and tracking library is called Jeeliz FaceFilter API. It handles multi-face detection, and for each tracked face it provides the rotation angles and the mouth opening factor. It is perfect to build your own Snapchat/MSQRD like face filters running in the browser. It comes with dozen of integration demo, including a face swap.

References