labsfilter

Labsfilter

Labsfilter is a module to easily and quickly render 3D filters on your face with your webcam in real-time.

Installation

npm i labsfilter

Import

import labsfilter from './node_modules/labsfilter/index.js'

Functions

labsfilter.init(<Object> settings, <String> texturepath, <String> modelpath);

Inits everything to start the filter applying later The texturepath and modelpath are optional but note that the labsfilter.start() function won't work before you specify the two path's with either the labsfilter.init() or labsfilter.setFilterModel() function
the settings object can contains the following items

• video: videoElement //HTML video element for webcam usage Required
• canvas: canvasElement //HTML canvas element for filter applying Required
• NCCpath: path //Path to the neural network Optional 'NNC.json' by default
• facingMode: 'user' //Defines which camera will be used Optional 'user' by default

labsfilter.start(); Starts the face detector and filter applying.

labsfilter.setFilterModel(<String> texturepath, <String> modelpath); Sets or updates the 3D filter based on the given texture and 3D model. The texturepath should always be and image (.png, .jpg etc) and the modalpath should always be an .obj file

labsfilter.stop(); Completely stops the face detector ands removes the 3D filter

Neural networks

Available neural networks for the face detector:

• NNC.json
• NNCwideAngles.json
• NNClight.json
• NNCveryLight.json
• NNCviewTop.json
• NNCdeprecated.json
• NNCdeprecated.json
• NNCIntel1536.json
• NNCNNC4Expr0.json

Sample setup

import labsfilter from './node_modules/labsfilter/index.js'
let webcam = document.querySelector('#webcam'); //HTML video element
let canvas = document.querySelector('canvas'); //HTML canvas element

let texture = 'models/tiger/tiger.png'; //Path to the texture for the model (.png, .jpg etc.)
let model = 'models/tiger/tiger.obj'; //Path to the 3D model (must be .obj file)

let settings = {video: webcam, canvas: canvas, facingMode: 'user'};

labsfilter.init(settings, texture, model); //Initializes everything
labsfilter.start(); //Starts the face detector and 3D filter applying.

Demo

Live demo | Source code

License

ISC