Reference version

SDK 50 (latest)

Archive Expo Snack Discord and Forums Newsletter

Expo FaceDetector

GitHub

npm

A library that uses Google Mobile Vision to detect faces on images.

Deprecated: This library will no longer be available from SDK 51. We recommend react-native-vision-camera if you require this functionality.

expo-face-detector lets you use the power of the Google's ML Kit framework to detect faces on images.

Platform Compatibility

Android Device	Android Emulator	iOS Device	iOS Simulator	Web

Known issues
Android

Face detector does not recognize faces that aren't aligned with the interface (top of the interface matches top of the head).

Installation

This module is not available in the Expo Go app because it has dependencies that break builds for iOS Simulators.

You can create a development build to work with this package.

Terminal

- npx expo install expo-face-detector

If you're installing this in a bare React Native app, you should also follow these additional installation instructions.

Usage

Settings

To configure detector's behavior modules pass a DetectionOptions object which is then interpreted by this module.

Example

You can use the following snippet to detect faces in a fast mode without detecting landmarks or whether a face is smiling.

Quick face detection

import * as React from 'react';
import { Camera } from 'expo-camera';
import * as FaceDetector from 'expo-face-detector';

const App = () => (
  <Camera
    // other props
    onFacesDetected={handleFacesDetected}
    faceDetectorSettings={{
      mode: FaceDetector.FaceDetectorMode.fast,
      detectLandmarks: FaceDetector.FaceDetectorLandmarks.none,
      runClassifications: FaceDetector.FaceDetectorClassifications.none,
      minDetectionInterval: 100,
      tracking: true,
    }}
  />
);

const handleFacesDetected = ({ faces }) => {
  console.log(faces);
};

export default App;

API

import * as FaceDetector from 'expo-face-detector';

Methods

Deprecated. If you require this functionality, we recommend using react-native-vision-camera

`FaceDetector.detectFacesAsync(uri, options)`

Name	Type	Description
uri	`string`	`file://` URI to the image.
options (optional)	`DetectionOptions`	A map of detection options. Default: `{}`

Detect faces on a picture.

Returns

Promise<DetectionResult>

Returns a Promise which fulfils with DetectionResult object.

Types

`DetectionOptions`

In order to configure detector's behavior modules pass a settings object which is then interpreted by this module.

Name	Type	Description
detectLandmarks (optional)	`FaceDetectorLandmarks`	Whether to detect and return landmarks positions on the face (ears, eyes, mouth, cheeks, nose). Use `FaceDetector.FaceDetectorLandmarks.{all, none}`.
minDetectionInterval (optional)	`number`	Minimal interval in milliseconds between two face detection events being submitted to JS. Use, when you expect lots of faces for long time and are afraid of JS Bridge being overloaded. Default: `0`
mode (optional)	`FaceDetectorMode`	Whether to detect faces in fast or accurate mode. Use `FaceDetector.FaceDetectorMode.{fast, accurate}`.
runClassifications (optional)	`FaceDetectorClassifications`	Whether to run additional classifications on detected faces (smiling probability, open eye probabilities). Use `FaceDetector.FaceDetectorClassifications.{all, none}`.
tracking (optional)	`boolean`	Flag to enable tracking of faces between frames. If true, each face will be returned with `faceID` attribute which should be consistent across frames. Default: `false`

`DetectionResult`

Name	Type	Description
faces	`FaceFeature[]`	Array of faces objects.
image	`Image`	-

`FaceFeature`

Name	Type	Description
bottomMouthPosition (optional)	`Point`	Position of the bottom edge of the mouth in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
bounds	`FaceFeatureBounds`	An object containing face bounds.
faceID (optional)	`number`	A face identifier (used for tracking, if the same face appears on consecutive frames it will have the same `faceID`).
leftCheekPosition (optional)	`Point`	Position of the left cheek in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
leftEarPosition (optional)	`Point`	Position of the left ear in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
leftEyeOpenProbability (optional)	`number`	Probability that the left eye is open. Returned only if detection classifications property is set to `FaceDetectorClassifications.all`.
leftEyePosition (optional)	`Point`	Position of the left eye in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
leftMouthPosition (optional)	`Point`	Position of the left edge of the mouth in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
mouthPosition (optional)	`Point`	Position of the center of the mouth in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
noseBasePosition (optional)	`Point`	Position of the nose base in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
rightCheekPosition (optional)	`Point`	Position of the right cheek in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
rightEarPosition (optional)	`Point`	Position of the right ear in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
rightEyeOpenProbability (optional)	`number`	Probability that the right eye is open. Returned only if detection classifications property is set to `FaceDetectorClassifications.all`.
rightEyePosition (optional)	`Point`	Position of the right eye in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
rightMouthPosition (optional)	`Point`	Position of the right edge of the mouth in image coordinates. Returned only if detection classifications property is set to `FaceDetectorLandmarks.all`.
rollAngle (optional)	`number`	Roll angle of the face (bank).
smilingProbability (optional)	`number`	Probability that the face is smiling. Returned only if detection classifications property is set to `FaceDetectorClassifications.all`.
yawAngle (optional)	`number`	Yaw angle of the face (heading, turning head left or right).

`FaceFeatureBounds`

Name	Type	Description
origin	`Point`	Position of the top left corner of a square containing the face in image coordinates,
size	`{ height: number, width: number }`	Size of the square containing the face in image coordinates,

`Image`

Name	Type	Description
height	`number`	Height of the image in pixels.
orientation	`number`	Orientation of the image (value conforms to the EXIF orientation tag standard).
uri	`string`	URI of the image.
width	`number`	Width of the image in pixels.

`Point`

Name	Type	Description
x	`number`	-
y	`number`	-

Expo FaceDetector

Platform Compatibility

Known issues
Android

Installation

Usage

Settings

Example

API

Methods

`FaceDetector.detectFacesAsync(uri, options)`

Types

`DetectionOptions`

`DetectionResult`

`FaceFeature`

`FaceFeatureBounds`

`Image`

`Point`

Enums

`FaceDetectorClassifications`

`none`

`all`

`FaceDetectorLandmarks`

`none`

`all`

`FaceDetectorMode`

`fast`

`accurate`