Capacitor ML Kit Face Detection Plugin

Unofficial Capacitor plugin for ML Kit Face Detection.[^1]

Use Cases

The Face Detection plugin is typically used whenever an app needs to locate faces in an image, for example:

• Photo organization: Detect which photos contain faces and where they are located, for example to crop thumbnails around them.
• Selfie effects: Use the detected face landmarks and contours to position stickers or other virtual elements on faces.
• Smile detection: Use the classification results to determine the probability that a person is smiling or has their eyes open.
• Face tracking: Maintain a consistent ID for each face across consecutive frames using the tracking option.

Compatibility

| Plugin Version | Capacitor Version | Status | | -------------- | ----------------- | -------------- | | 8.x.x | >=8.x.x | Active support | | 7.x.x | 7.x.x | Deprecated | | 6.x.x | 6.x.x | Deprecated |

Installation

You can use our AI-Assisted Setup to install the plugin. Add the Capawesome Skills to your AI tool using the following command:

npx skills add capawesome-team/skills --skill capacitor-plugins

Then use the following prompt:

Use the `capacitor-plugins` skill from `capawesome-team/skills` to install the `@capacitor-mlkit/face-detection` plugin in my project.

If you prefer Manual Setup, install the plugin by running the following commands and follow the platform-specific instructions below:

npm install @capacitor-mlkit/face-detection
npx cap sync

Attention: This plugin only supports CocoaPods for iOS dependency management. Swift Package Manager (SPM) is not supported for the ML Kit SDK, see this comment.

Android

Dependencies

You need to add the following meta data in the application tag in your AndroidManifest.xml:

<meta-data android:name="com.google.mlkit.vision.DEPENDENCIES" android:value="face"/>
<!-- To use multiple models: android:value="face,model2,model3" -->

Variables

If needed, you can define the following project variable in your app’s variables.gradle file to change the default version of the dependency:

• $mlkitFaceDetectionVersion version of com.google.mlkit:face-detection (default: 16.1.7)
• $playServicesMlkitFaceDetectionVersion version of com.google.android.gms:play-services-mlkit-face-detection (default: 17.1.0)

This can be useful if you encounter dependency conflicts with other plugins in your project.

iOS

Minimum Deployment Target

Make sure to set the deployment target in your ios/App/Podfile to at least 15.5:

platform :ios, '15.5'

Configuration

No configuration required for this plugin.

Demo

A working example can be found here: robingenz/capacitor-mlkit-plugin-demo

Usage

The following example shows how to detect faces in an image.

Detect faces in an image

Detect human faces in an image at a local path. You can configure the accuracy/speed trade-off and enable landmarks, contours, classification (smiling and eyes open probabilities), and face tracking. Only available on Android and iOS:

import {
  FaceDetection,
  PerformanceMode,
  LandmarkMode,
  ContourMode,
  ClassificationMode,
} from '@capacitor-mlkit/face-detection';

const processImage = async () => {
  const { faces } = await FaceDetection.processImage({
    path: 'path/to/image.jpg',
    performanceMode: PerformanceMode.Fast,
    landmarkMode: LandmarkMode.All,
    contourMode: ContourMode.All,
    classificationMode: ClassificationMode.All,
    minFaceSize: 0.1,
    enableTracking: false,
  });
  return faces;
};

API

• processImage(...)
• Interfaces
• Enums

processImage(...)

processImage(options: ProcessImageOptions) => Promise<ProcessImageResult>

Detects human faces from the supplied image.

Only available on Android and iOS.

| Param | Type | | ------------- | ------------------------------------------------------------------- | | options | ProcessImageOptions |

Returns: Promise<ProcessImageResult>

Since: 5.1.0

Interfaces

ProcessImageResult

| Prop | Type | Description | Since | | ----------- | ------------------- | ------------------- | ----- | | faces | Face[] | The detected faces. | 5.1.0 |

Face

Represents a face detected by FaceDetector.

| Prop | Type | Description | Since | | ----------------------------- | ------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- | | bounds | Rect | Returns the axis-aligned bounding rectangle of the detected face. | 5.1.0 | | landmarks | FaceLandmark[] | Returns a list of face landmarks. | 5.1.0 | | contours | FaceContour[] | Returns a list of face contours. | 5.1.0 | | trackingId | number | Returns the tracking ID if the tracking is enabled. | 5.1.0 | | headEulerAngleX | number | Returns the rotation of the face about the horizontal axis of the image. Positive euler X is the face is looking up. | 5.1.0 | | headEulerAngleY | number | Returns the rotation of the face about the vertical axis of the image. Positive euler y is when the face turns toward the right side of the image that is being processed. | 5.1.0 | | headEulerAngleZ | number | Returns the rotation of the face about the axis pointing out of the image. Positive euler z is a counter-clockwise rotation within the image plane. | 5.1.0 | | smilingProbability | number | Returns a value between 0.0 and 1.0 giving a probability that the face is smiling. | 5.1.0 | | leftEyeOpenProbability | number | Returns a value between 0.0 and 1.0 giving a probability that the face's left eye is open. | 5.1.0 | | rightEyeOpenProbability | number | Returns a value between 0.0 and 1.0 giving a probability that the face's right eye is open. | 5.1.0 |

Rect

Rect holds four integer coordinates for a rectangle.

| Prop | Type | Description | Since | | ------------ | ------------------- | ---------------------------------------------------- | ----- | | left | number | The X coordinate of the left side of the rectangle. | 5.1.0 | | top | number | The Y coordinate of the top of the rectangle. | 5.1.0 | | right | number | The X coordinate of the right side of the rectangle. | 5.1.0 | | bottom | number | The Y coordinate of the bottom of the rectangle. | 5.1.0 |

FaceLandmark

Represent a face landmark. A landmark is a point on a detected face, such as an eye, nose, or mouth.

| Prop | Type | Description | Since | | -------------- | ----------------------------------------------------- | ------------------------------------------------------------------------------------------------ | ----- | | type | LandmarkType | Gets the FaceLandmark.LandmarkType type. | 5.1.0 | | position | Point | Gets a 2D point for landmark position, where (0, 0) is the upper-left corner of the image. | 5.1.0 |

Point

Point holds two coordinates

| Prop | Type | | ------- | ------------------- | | x | number | | y | number |

FaceContour

Represent a face contour. A contour is a list of points on a detected face, such as the mouth.

| Prop | Type | Description | Since | | ------------ | --------------------------------------------------- | --------------------------------------------------------------------------------------------------- | ----- | | type | ContourType | Gets the FaceContour.ContourType type. | 5.1.0 | | points | Point[] | Gets a list of 2D points for this face contour, where (0, 0) is the upper-left corner of the image. | 5.1.0 |

ProcessImageOptions

| Prop | Type | Description | Default | Since | | ------------------------ | ----------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------ | ----- | | path | string | The local path to the image file. | | 5.1.0 | | performanceMode | PerformanceMode | Defines options to control accuracy / speed trade-offs in performing face detection. | PerformanceMode.Fast | 5.1.0 | | landmarkMode | LandmarkMode | Defines options to enable face landmarks or not. | LandmarkMode.None | 5.1.0 | | contourMode | ContourMode | Defines options to enable face contours or not. | ContourMode.None | 5.1.0 | | classificationMode | ClassificationMode | Defines options for characterizing attributes such as "smiling" * and "eyes open". | ClassificationMode.None | 5.1.0 | | minFaceSize | number | Sets the smallest desired face size, expressed as a proportion of the width of the head to the image width. | 0.1 | 5.1.0 | | enableTracking | boolean | Enables face tracking, which will maintain a consistent ID for each face when processing consecutive frames. Tracking should be disabled for handling a series of non-consecutive still images. | false | 5.1.0 |

Enums

LandmarkType

| Members | Value | Description | Since | | ----------------- | --------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- | | MouthBottom | 0 | The center of the subject's bottom lip. | 5.1.0 | | LeftCheek | 1 | The midpoint between the subject's left mouth corner and the outer corner of the subject's left eye. For full profile faces, this becomes the centroid of the nose base, nose tip, left ear lobe and left ear tip. | 5.1.0 | | LeftEar | 3 | The midpoint of the subject's left ear tip and left ear lobe. | 5.1.0 | | LeftEye | 4 | The center of the subject's left eye cavity. | 5.1.0 | | MouthLeft | 5 | The subject's left mouth corner where the lips meet. | 5.1.0 | | NoseBase | 6 | The midpoint between the subject's nostrils where the nose meets the face. | 5.1.0 | | RightCheek | 7 | The midpoint between the subject's right mouth corner and the outer corner of the subject's right eye. For full profile faces, this becomes the centroid of the nose base, nose tip, right ear lobe and right ear tip. | 5.1.0 | | RightEar | 9 | The midpoint of the subject's right ear tip and right ear lobe. | 5.1.0 | | RightEye | 10 | The center of the subject's right eye cavity. | 5.1.0 | | MouthRight | 11 | The subject's right mouth corner where the lips meet. | 5.1.0 |

ContourType

| Members | Value | Description | Since | | ------------------------ | --------------- | -------------------------------------------------- | ----- | | Face | 1 | The outline of the subject's face. | 5.1.0 | | LeftEyebrowTop | 2 | The top outline of the subject's left eyebrow. | 5.1.0 | | LeftEyebrowBottom | 3 | The bottom outline of the subject's left eyebrow. | 5.1.0 | | RightEyebrowTop | 4 | The top outline of the subject's right eyebrow. | 5.1.0 | | RightEyebrowBottom | 5 | The bottom outline of the subject's right eyebrow. | 5.1.0 | | LeftEye | 6 | The outline of the subject's left eye cavity. | 5.1.0 | | RightEye | 7 | The outline of the subject's right eye cavity. | 5.1.0 | | UpperLipTop | 8 | The top outline of the subject's upper lip. | 5.1.0 | | UpperLipBottom | 9 | The bottom outline of the subject's upper lip. | 5.1.0 | | LowerLipTop | 10 | The top outline of the subject's lower lip. | 5.1.0 | | LowerLipBottom | 11 | The bottom outline of the subject's lower lip. | 5.1.0 | | NoseBridge | 12 | The outline of the subject's nose bridge. | 5.1.0 | | NoseBottom | 13 | The outline of the subject's nose bridge. | 5.1.0 | | LeftCheek | 14 | The center of the left cheek. | 5.1.0 | | RightCheek | 15 | The center of the right cheek. | 5.1.0 |

PerformanceMode

| Members | Value | Description | Since | | -------------- | -------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- | | Fast | 1 | Indicates a preference for speed in the options that may make an accuracy vs. speed trade-off. This will tend to detect fewer faces and may be less precise in determining values such as position, but will run faster. | 5.1.0 | | Accurate | 2 | Indicates a preference for accuracy in the options that may make an accuracy vs. speed trade-off. This will tend to detect more faces and may be more precise in determining values such as position, at the cost of speed. | 5.1.0 |

LandmarkMode

| Members | Value | Description | Since | | ---------- | -------------- | ------------------------------------------------------------------ | ----- | | None | 1 | Does not perform landmark detection. | 5.1.0 | | All | 2 | Detects FaceLandmark for a given face. | 5.1.0 |

ContourMode

| Members | Value | Description | Since | | ---------- | -------------- | --------------------------------------------------------------------------------------------------------------------- | ----- | | None | 1 | Does not perform contour detection. | 5.1.0 | | All | 2 | Detects FaceContour for a given face. Note that it would return contours for up to 5 faces | 5.1.0 |

ClassificationMode

| Members | Value | Description | Since | | ---------- | -------------- | -------------------------------------------------- | ----- | | None | 1 | Does not perform classification. | 5.1.0 | | All | 2 | Performs "eyes open" and "smiling" classification. | 5.1.0 |

FAQ

Which platforms are supported by this plugin?

The processImage(...) method is only available on Android and iOS. The Web platform is not supported by the underlying ML Kit Face Detection SDK.

Does the plugin recognize who is in the picture?

No, the plugin performs face detection, not face recognition. It detects the position of faces in an image along with attributes such as landmarks, contours, head rotation, and smiling probability, but it does not identify individuals. With the enableTracking option you can only maintain a consistent ID for the same face across consecutive frames.

What is the difference between landmarks and contours?

A landmark is a single point on a detected face, such as an eye, nose, or mouth. A contour is a list of points that outlines a facial feature, such as the face oval, eyebrows, eyes, or lips. Landmarks are enabled via the landmarkMode option and contours via the contourMode option. Note that contours are only returned for up to 5 faces.

How can I detect whether a person is smiling or has their eyes open?

Set the classificationMode option to All. The detected faces then contain the smilingProbability, leftEyeOpenProbability, and rightEyeOpenProbability properties, each a value between 0.0 and 1.0.

What is the difference between this plugin and the Face Mesh Detection plugin?

This plugin detects faces with bounding boxes, landmarks, contours, and classifications on Android and iOS. The ML Kit Face Mesh Detection plugin returns a dense mesh of 468 3D points and triangle information for detected faces, but is only available on Android.

Can I use this plugin with Ionic, React, Vue or Angular?

Yes, the plugin is framework-agnostic. It works in any Capacitor app regardless of the web framework, including Ionic with Angular, React, or Vue, as well as plain JavaScript projects.

Related Plugins

• ML Kit Face Mesh Detection: Detect face meshes with 468 3D points in images.
• ML Kit Selfie Segmentation: Segment selfies from the background with ML Kit Selfie Segmentation.
• ML Kit Subject Segmentation: Separate subjects from the background with ML Kit Subject Segmentation.
• ML Kit Pose Detection: Detect body poses alongside faces in the same scene.

Terms & Privacy

This plugin uses the Google ML Kit:

• Terms & Privacy
• Android Data Disclosure
• iOS Data Disclosure

Newsletter

Stay up to date with the latest news and updates about the Capawesome, Capacitor, and Ionic ecosystem by subscribing to our Capawesome Newsletter.

Changelog

See CHANGELOG.md.

License

See LICENSE.

[^1]: This project is not affiliated with, endorsed by, sponsored by, or approved by Google LLC or any of their affiliates or subsidiaries.