Viola

Viola android face detection library detects faces automatically from a bitmap, crop faces using the predefined algorithms, and provides supplementary information of the detected faces.

Features:
-> Prominent face detection
-> Multiple face detection
-> Automatic face crop
-> Face classification information
-> Age classification support

Getting Started

Demo application

Download sample application sample_v_1.1.1

Compatibility

Minimum Android SDK : Viola 1.1.3 requires a minimum API level of 21.

Installation

Download the latest aar from jitpack or grab via

root build.gradle:

allprojects {
    repositories {
        ...
        maven { url 'https://jitpack.io' }
    }
}

app build.gradle:

dependencies {
  implementation 'com.github.darwinfrancis:viola:1.1.4'
}

or Maven:

<dependency>
  <groupId>com.github.darwinfrancis</groupId>
  <artifactId>viola</artifactId>
  <version>1.1.4</version>
  <type>pom</type>
</dependency>

Usage

Kotlin

val viola = Viola(listener)
viola.addAgeClassificationPlugin(context) //optional, available via external dependency

viola.detectFace(bitmap)

private val listener: FaceDetectionListener = object : FaceDetectionListener {

        override fun onFaceDetected(result: Result) {}

        override fun onFaceDetectionFailed(error: FaceDetectionError, message: String) {}
}

with FaceOptions

val faceOption =
    FaceOptions.Builder()
               .enableProminentFaceDetection()
               .enableDebug()
               .build()
viola.detectFace(bitmap,faceOption)

Java

Viola viola = new Viola(listener);
viola.addAgeClassificationPlugin(context) //optional, available via external dependency

viola.detectFace(bitmap);

private final FaceDetectionListener listener = new FaceDetectionListener() {
        @Override
        public void onFaceDetected(@NotNull Result result) { }

        @Override
        public void onFaceDetectionFailed(@NotNull FaceDetectionError error, @NotNull String message) { }
};

with FaceOptions

FaceOptions faceOptions = new FaceOptions.Builder()
                .enableProminentFaceDetection()
                .enableDebug()
                .build();
viola.detectFace(bitmap,faceOptions);

Configure the face detector

Viola is currently extended with the following configurations. Instructions on how to use them in your own application are linked below.

FaceOptions	Description	Type	Default
`prominentFaceDetection`	Indicates whether to detect all faces, or to only detect the most prominent face	boolean	false
`cropAlgorithm`	Extended option for controlling crop constraints	CropAlgorithm	CropAlgorithm.THREE_BY_FOUR
`minFaceSize`	The minimum size percentage, relative to the image, of faces to detect	int	15
`ageClassification`	Indicate whether to enable age classification on detected face	boolean	false
`debug`	enables debug log	boolean	false

CropAlgorithm.THREE_BY_FOUR : Performs face crop in three by four ratio

CropAlgorithm.SQUARE : Performs face crop in 1:1 ratio(useful when showing face in circular view)

CropAlgorithm.LEAST : Performs face crop with minimum padding(possibly reduce face overlap on multi face detection)

NOTE : add viola-age dependency to your application build.gradle file and viola.addAgeClassificationPlugin(this) to your viola sdk instance for ageClassification feature to work on detected face.

Face detection result

Viola provides the following values in Result class

Result	Description	Type
`faceCount`	The number of faces croped	int
`facePortraits`	Contains list of cropped faces	FacePortrait
`face`	The cropped face bitmap	Bitmap
`smileProbability`	Giving a probability that the face is smiling	Float
`leftEyeOpenProbability`	Giving a probability that the face's left eye is open	Float
`rightEyeOpenProbability`	Giving a probability that the face's right eye is open	Float
`pixelBetweenEyes`	The number of pixels between the eyes	Double
`faceSizePercentage`	The size of face relative to the input image	Float
`facePose`	Provides rotation of face in X,Y,Z plane	FacePose
`ageRange`	Provides age range of detected face(s)	String

Author

Darwin Francis - @darwinfrancis on GitHub, @darwin-francis on linkedin

Roadmap

-Add gender classification feature as plugin
-Add emotion classification feature as plugin
-Implement face detection and cropping from live camera preview.

Other works

Viola-Age age classification library for android.

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

License

MIT License

Copyright (c) 2020 Darwin Francis

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

darwinfrancis/viola-face-detection