@vladmandic/face-api

FaceAPI

Note

This is updated face-api.js with latest available TensorFlow/JS as the original face-api.js is not compatible with tfjs 2.0+.

Forked from face-api.js version 0.22.2 released on March 22nd, 2020

Currently based on TensorFlow/JS 2.8.0

Why?

Because I needed Face-API that does not cause version conflict with newer TFJS 2.0 that I use accross my projects
And since original Face-API was open-source, I've released this version as well

Unfortunately, changes ended up being too large for a simple pull request on original Face-API and it ended up being a full-fledged version on its own

Differences

• Compatible with TensorFlow/JS 2.0+
• Compatible with WebGL, CPU and WASM TFJS Browser backends
• Compatible with both tfjs-node and tfjs-node-gpu TFJS NodeJS backends
• Updated all type castings for TypeScript type checking to TypeScript 4.1
• Switched bundling from UMD to ESM + CommonJS with fallback to IIFE
  Resulting code is optimized per-platform instead of being universal
  Fully tree shakable when imported as an ESM module
  Browser bundle process uses ESBuild instead of Rollup
• Typescript build process now targets ES2018 and instead of dual ES5/ES6
  Resulting code is clean ES2018 JavaScript without polyfills
• Removed old tests, docs, examples
• Removed old package dependencies (karma, jasmine, babel, etc.)
• Updated all package dependencies
• Updated TensorFlow/JS dependencies since backends were removed from @tensorflow/tfjs-core
• Updated mobileNetv1 model due to batchNorm() dependency
• Added version class that returns JSON object with version of FaceAPI as well as linked TFJS
• Removed mtcnn and tinyYolov2 models as they were non-functional in latest public version of Face-API
  If there is a demand, I can re-implement them back.

Which means valid models are tinyFaceDetector and mobileNetv1

---

Examples

Browser

Browser example that uses both models as well as all of the extensions is included in /example/index.html
Example can be accessed directly using Git pages using URL: https://vladmandic.github.io/face-api/example/

Note: Photos shown below are taken by me

NodeJS

Two NodeJS examples are:

• /example/node-singleprocess.js: Regular usage of FaceAPI from NodeJS
• /example/node-multiprocess.js: Multiprocessing showcase that uses pool of worker processes (node-multiprocess-worker.js Main starts fixed pool of worker processes with each worker having it's instance of FaceAPI
  Workers communicate with main when they are ready and main dispaches job to each ready worker until job queue is empty

2020-12-08 08:30:01 INFO:  @vladmandic/face-api version 0.9.1
2020-12-08 08:30:01 INFO:  User: vlado Platform: linux Arch: x64 Node: v15.0.1
2020-12-08 08:30:01 INFO:  FaceAPI multi-process test
2020-12-08 08:30:01 STATE:  Main: started worker: 265238
2020-12-08 08:30:01 STATE:  Main: started worker: 265244
2020-12-08 08:30:02 STATE:  Worker: PID: 265238 TensorFlow/JS 2.7.0 FaceAPI 0.9.1 Backend: tensorflow
2020-12-08 08:30:02 STATE:  Worker: PID: 265244 TensorFlow/JS 2.7.0 FaceAPI 0.9.1 Backend: tensorflow
2020-12-08 08:30:02 STATE:  Main: dispatching to worker: 265238
2020-12-08 08:30:02 STATE:  Main: dispatching to worker: 265244
2020-12-08 08:30:02 DATA:  Worker received message: 265238 { image: 'example/sample (1).jpg' }
2020-12-08 08:30:02 DATA:  Worker received message: 265244 { image: 'example/sample (2).jpg' }
2020-12-08 08:30:04 DATA:  Main: worker finished: 265238 detected faces: 3
2020-12-08 08:30:04 STATE:  Main: dispatching to worker: 265238
2020-12-08 08:30:04 DATA:  Main: worker finished: 265244 detected faces: 3
2020-12-08 08:30:04 STATE:  Main: dispatching to worker: 265244
2020-12-08 08:30:04 DATA:  Worker received message: 265238 { image: 'example/sample (3).jpg' }
2020-12-08 08:30:04 DATA:  Worker received message: 265244 { image: 'example/sample (4).jpg' }
2020-12-08 08:30:06 DATA:  Main: worker finished: 265238 detected faces: 3
2020-12-08 08:30:06 STATE:  Main: dispatching to worker: 265238
2020-12-08 08:30:06 DATA:  Worker received message: 265238 { image: 'example/sample (5).jpg' }
2020-12-08 08:30:06 DATA:  Main: worker finished: 265244 detected faces: 4
2020-12-08 08:30:06 STATE:  Main: dispatching to worker: 265244
2020-12-08 08:30:06 DATA:  Worker received message: 265244 { image: 'example/sample (6).jpg' }
2020-12-08 08:30:07 DATA:  Main: worker finished: 265238 detected faces: 5
2020-12-08 08:30:07 STATE:  Main: worker exit: 265238 0
2020-12-08 08:30:08 DATA:  Main: worker finished: 265244 detected faces: 4
2020-12-08 08:30:08 INFO:  Processed 12 images in 6826 ms
2020-12-08 08:30:08 STATE:  Main: worker exit: 265244 0

Note that @tensorflow/tfjs-node or @tensorflow/tfjs-node-gpu must be installed before using NodeJS example

---

Installation

Face-API ships with several pre-build versions of the library:

• dist/face-api.js: IIFE format for client-side Browser execution with TFJS pre-bundled
• dist/face-api.esm.js: ESM format for client-side Browser execution with TFJS pre-bundled
• dist/face-api.esm-nobundle.js: ESM format for client-side Browser execution without TFJS pre-bundled
• dist/face-api.node.js: CommonJS format for server-side NodeJS execution without TFJS pre-bundled
• dist/face-api.node-gpu.js: CommonJS format for server-side NodeJS execution without TFJS pre-bundled and optimized for CUDA GPU acceleration

Defaults are:

{
  "main": "dist/face-api.node-js",
  "module": "dist/face-api.esm.js",
  "browser": "dist/face-api.esm.js",
}

Bundled TFJS can be used directly via export: faceapi.tf

Reason for additional nobundle version is if you want to include a specific version of TFJS and not rely on pre-packaged one

FaceAPI is compatible with TFJS 2.0+

All versions include sourcemap and asset manifest

---

There are several ways to use Face-API:

1. IIFE script

Recommened for quick tests and backward compatibility with older Browsers that do not support ESM such as IE

This is simplest way for usage within Browser
Simply download dist/face-api.js, include it in your HTML file & it's ready to use

<script src="dist/face-api.js"><script>

IIFE script bundles TFJS and auto-registers global namespace faceapi within Window object which can be accessed directly from a <script> tag or from your JS file.

2. ESM module

Recommended for usage within Browser

2.1. Direct Import

To use ESM import directly in a Browser, you must import your script (e.g. index.js) with a type="module"

  <script src="./index.js" type="module">

and then in your index.js

  import * as faceapi from 'dist/face-api.esm.js';

2.2. With Bundler

Same as above, but expectation is that you've installed @vladmandic/faceapi package:

  npm install @vladmandic/face-api 

and that you'll package your application using a bundler such as webpack, rollup or esbuild
in which case, you do not need to import a script as module - that depends on your bundler configuration

  import * as faceapi from '@vladmandic/face-api';

or if your bundler doesn't recognize recommended type, force usage with:

  import * as faceapi from '@vladmandic/face-api/dist/face-api.esm.js';

or to use non-bundled version

  import * as tf from `@tensorflow/tfjs`;
  import * as faceapi from '@vladmandic/face-api/dist/face-api.esm-nobundle.js';

3. NPM module

3.1. Import CommonJS

Recommended for NodeJS projects

Node: Face-API for NodeJS does not bundle TFJS due to binary dependencies that are installed during TFJS installation

Install with:

  npm install @tensorflow/tfjs-node
  npm install @vladmandic/face-api 

And then use with:

  const tf = require('@tensorflow/tfjs-node')
  const faceapi = require('@vladmandic/face-api');

If you want to force CommonJS module instead of relying on recommended field:

  const faceapi = require('@vladmandic/face-api/dist/face-api.node.js');

If you want to GPU Accelerated execution in NodeJS, you must have CUDA libraries already installed and working
Then install appropriate version of Face-API:

  npm install @tensorflow/tfjs-node
  npm install @vladmandic/face-api 

And then use with:

  const tf = require('@tensorflow/tfjs-node-gpu')
  const faceapi = require('@vladmandic/face-api/dist/face-api.node-gpu.js'); // this loads face-api version with correct bindings for tfjs-node-gpu

---

Weights

Pretrained models and their weights are includes in ./model.

Build

If you want to do a full rebuild, either download npm module

npm install @vladmandic/face-api
cd node_modules/@vladmandic/face-api

or clone a git project

git clone https://github.com/vladmandic/face-api
cd face-api

Then install all dependencies and run rebuild:

npm install
npm run build

Build process uses script build.js that creates optimized build for each target:

npm run build

> @vladmandic/face-api@0.8.9 build /home/vlado/dev/face-api
> rimraf dist/* && node ./build.js

2020-12-02 16:31:23 INFO:  @vladmandic/face-api version 0.8.9
2020-12-02 16:31:23 INFO:  User: vlado Platform: linux Arch: x64 Node: v15.0.1
2020-12-02 16:31:23 INFO:  Build: file startup all target: es2018
2020-12-02 16:31:23 STATE:  Build for: node type: tfjs: { imports: 1, importBytes: 39, outputBytes: 1042, outputFiles: 'dist/tfjs.esm.js' }
2020-12-02 16:31:23 STATE:  Build for: node type: node: { imports: 160, importBytes: 228038, outputBytes: 134190, outputFiles: 'dist/face-api.node.js' }
2020-12-02 16:31:23 STATE:  Build for: nodeGPU type: tfjs: { imports: 1, importBytes: 43, outputBytes: 1046, outputFiles: 'dist/tfjs.esm.js' }
2020-12-02 16:31:23 STATE:  Build for: nodeGPU type: node: { imports: 160, importBytes: 228042, outputBytes: 134198, outputFiles: 'dist/face-api.node-gpu.js' }
2020-12-02 16:31:23 STATE:  Build for: browserNoBundle type: tfjs: { imports: 1, importBytes: 1784, outputBytes: 244, outputFiles: 'dist/tfjs.esm.js' }
2020-12-02 16:31:23 STATE:  Build for: browserNoBundle type: esm: { imports: 160, importBytes: 227240, outputBytes: 131024, outputFiles: 'dist/face-api.esm-nobundle.js' }
2020-12-02 16:31:24 STATE:  Build for: browserBundle type: tfjs: { modules: 1045, moduleBytes: 3718721, imports: 7, importBytes: 1784, outputBytes: 1501677, outputFiles: 'dist/tfjs.esm.js' }
2020-12-02 16:31:24 STATE:  Build for: browserBundle type: iife: { imports: 162, importBytes: 1728673, modules: 576, moduleBytes: 1359851, outputBytes: 1903311, outputFiles: 'dist/face-api.js' }
2020-12-02 16:31:25 STATE:  Build for: browserBundle type: esm: { imports: 162, importBytes: 1728673, modules: 576, moduleBytes: 1359851, outputBytes: 1900836, outputFiles: 'dist/face-api.esm.js' }

---

Credits & Documentation

• Original project and usage documentation: Face-API
• Original model weighs: Face-API
• ML API Documentation: Tensorflow/JS

Changelog

0.9.5 2020/12/19 [e-mail]

• added tsc build typings