使用机器学习套件检测人脸网格信息 (Android)

您可以使用机器学习套件检测自拍类图片和视频中的人脸。

Face Mesh Detection API
SDK 名称face-mesh-detection
实现代码和资源会在构建时静态链接到您的应用。
对应用大小的影响~6.4MB
性能在大多数设备上为实时。

试试看

准备工作

  1. 请务必在项目级 build.gradle 文件中的 buildscript 和 allprojects 部分添加 Google 的 Maven 代码库。

  2. 将机器学习套件人脸网格检测库的依赖项添加到模块的应用级 Gradle 文件(通常为 app/build.gradle):

    dependencies {
     // ...
    
     implementation 'com.google.mlkit:face-mesh-detection:16.0.0-beta1'
    }
    

输入图片准则

  1. 拍摄图片时,人脸应距离设备摄像头约 2 米(7 英尺),以便人脸足够大,以实现最佳的人脸网格识别效果。一般来说,面部越大,面部网格识别效果就越好。

  2. 人脸应朝向摄像头,至少要露出半张脸。 人脸和摄像头之间有任何大型物体都可能会导致准确性降低。

如果您想在实时应用中检测人脸,还应考虑输入图片的整体尺寸。较小图片的处理速度相对较快,因此,以较低的分辨率捕获图片可以缩短延迟时间。不过,请牢记上述准确性要求,并确保正文的面部在图片中占尽可能大的部分。

配置人脸网格检测器

如果您想更改人脸网格检测器的任何默认设置,请使用 FaceMeshDetectorOptions 对象指定这些设置。您可以更改以下设置:

  1. setUseCase

    • BOUNDING_BOX_ONLY:仅为检测到的面部网格提供边界框。这是最快的人脸检测器,但有范围限制(人脸必须在摄像头附近 2 米 [7 英尺] 以内)。

    • FACE_MESH(默认选项):提供边界框和其他面网格信息(468 个 3D 点和三角形信息)。与 BOUNDING_BOX_ONLY 用例相比,延迟时间增加了约 15%,如在 Pixel 3 上测得。

例如:

Kotlin

val defaultDetector = FaceMeshDetection.getClient(
  FaceMeshDetectorOptions.DEFAULT_OPTIONS)

val boundingBoxDetector = FaceMeshDetection.getClient(
  FaceMeshDetectorOptions.Builder()
    .setUseCase(UseCase.BOUNDING_BOX_ONLY)
    .build()
)

Java

FaceMeshDetector defaultDetector =
        FaceMeshDetection.getClient(
                FaceMeshDetectorOptions.DEFAULT_OPTIONS);

FaceMeshDetector boundingBoxDetector = FaceMeshDetection.getClient(
        new FaceMeshDetectorOptions.Builder()
                .setUseCase(UseCase.BOUNDING_BOX_ONLY)
                .build()
        );

准备输入图片

如需检测图片中的人脸,请基于设备上的以下资源创建一个 InputImage 对象:Bitmapmedia.ImageByteBuffer、字节数组或文件。然后,将 InputImage 对象传递给 FaceDetectorprocess 方法。

对于人脸网格检测,您使用的图片尺寸应至少为 480x360 像素。如果您要实时检测人脸,以此最低分辨率捕获帧有助于减少延迟时间。

您可以基于不同来源创建 InputImage 对象,下文分别介绍了具体方法。

使用 media.Image

如需基于 media.Image 对象创建 InputImage 对象(例如从设备的相机捕获图片时),请将 media.Image 对象和图片的旋转角度传递给 InputImage.fromMediaImage()

如果您使用 CameraX 库,OnImageCapturedListenerImageAnalysis.Analyzer 类会为您计算旋转角度值。

Kotlin

private class YourImageAnalyzer : ImageAnalysis.Analyzer {

    override fun analyze(imageProxy: ImageProxy) {
        val mediaImage = imageProxy.image
        if (mediaImage != null) {
            val image = InputImage.fromMediaImage(mediaImage, imageProxy.imageInfo.rotationDegrees)
            // Pass image to an ML Kit Vision API
            // ...
        }
    }
}

Java

private class YourAnalyzer implements ImageAnalysis.Analyzer {

    @Override
    public void analyze(ImageProxy imageProxy) {
        Image mediaImage = imageProxy.getImage();
        if (mediaImage != null) {
          InputImage image =
                InputImage.fromMediaImage(mediaImage, imageProxy.getImageInfo().getRotationDegrees());
          // Pass image to an ML Kit Vision API
          // ...
        }
    }
}

如果您不使用可提供图片旋转角度的相机库,则可以根据设备的旋转角度和设备中相机传感器的朝向来计算旋转角度:

Kotlin

private val ORIENTATIONS = SparseIntArray()

init {
    ORIENTATIONS.append(Surface.ROTATION_0, 0)
    ORIENTATIONS.append(Surface.ROTATION_90, 90)
    ORIENTATIONS.append(Surface.ROTATION_180, 180)
    ORIENTATIONS.append(Surface.ROTATION_270, 270)
}

/**
 * Get the angle by which an image must be rotated given the device's current
 * orientation.
 */
@RequiresApi(api = Build.VERSION_CODES.LOLLIPOP)
@Throws(CameraAccessException::class)
private fun getRotationCompensation(cameraId: String, activity: Activity, isFrontFacing: Boolean): Int {
    // Get the device's current rotation relative to its "native" orientation.
    // Then, from the ORIENTATIONS table, look up the angle the image must be
    // rotated to compensate for the device's rotation.
    val deviceRotation = activity.windowManager.defaultDisplay.rotation
    var rotationCompensation = ORIENTATIONS.get(deviceRotation)

    // Get the device's sensor orientation.
    val cameraManager = activity.getSystemService(CAMERA_SERVICE) as CameraManager
    val sensorOrientation = cameraManager
            .getCameraCharacteristics(cameraId)
            .get(CameraCharacteristics.SENSOR_ORIENTATION)!!

    if (isFrontFacing) {
        rotationCompensation = (sensorOrientation + rotationCompensation) % 360
    } else { // back-facing
        rotationCompensation = (sensorOrientation - rotationCompensation + 360) % 360
    }
    return rotationCompensation
}

Java

private static final SparseIntArray ORIENTATIONS = new SparseIntArray();
static {
    ORIENTATIONS.append(Surface.ROTATION_0, 0);
    ORIENTATIONS.append(Surface.ROTATION_90, 90);
    ORIENTATIONS.append(Surface.ROTATION_180, 180);
    ORIENTATIONS.append(Surface.ROTATION_270, 270);
}

/**
 * Get the angle by which an image must be rotated given the device's current
 * orientation.
 */
@RequiresApi(api = Build.VERSION_CODES.LOLLIPOP)
private int getRotationCompensation(String cameraId, Activity activity, boolean isFrontFacing)
        throws CameraAccessException {
    // Get the device's current rotation relative to its "native" orientation.
    // Then, from the ORIENTATIONS table, look up the angle the image must be
    // rotated to compensate for the device's rotation.
    int deviceRotation = activity.getWindowManager().getDefaultDisplay().getRotation();
    int rotationCompensation = ORIENTATIONS.get(deviceRotation);

    // Get the device's sensor orientation.
    CameraManager cameraManager = (CameraManager) activity.getSystemService(CAMERA_SERVICE);
    int sensorOrientation = cameraManager
            .getCameraCharacteristics(cameraId)
            .get(CameraCharacteristics.SENSOR_ORIENTATION);

    if (isFrontFacing) {
        rotationCompensation = (sensorOrientation + rotationCompensation) % 360;
    } else { // back-facing
        rotationCompensation = (sensorOrientation - rotationCompensation + 360) % 360;
    }
    return rotationCompensation;
}

然后,将 media.Image 对象及其旋转角度值传递给 InputImage.fromMediaImage()

Kotlin

val image = InputImage.fromMediaImage(mediaImage, rotation)

Java

InputImage image = InputImage.fromMediaImage(mediaImage, rotation);

使用文件 URI

如需基于文件 URI 创建 InputImage 对象,请将应用上下文和文件 URI 传递给 InputImage.fromFilePath()。如果您使用 ACTION_GET_CONTENT intent 提示用户从图库应用中选择图片,则这一操作非常有用。

Kotlin

val image: InputImage
try {
    image = InputImage.fromFilePath(context, uri)
} catch (e: IOException) {
    e.printStackTrace()
}

Java

InputImage image;
try {
    image = InputImage.fromFilePath(context, uri);
} catch (IOException e) {
    e.printStackTrace();
}

使用 ByteBufferByteArray

如需基于 ByteBufferByteArray 创建 InputImage 对象,请先按先前 media.Image 输入的说明计算图片旋转角度。然后,使用缓冲区或数组以及图片的高度、宽度、颜色编码格式和旋转角度创建 InputImage 对象:

Kotlin

val image = InputImage.fromByteBuffer(
        byteBuffer,
        /* image width */ 480,
        /* image height */ 360,
        rotationDegrees,
        InputImage.IMAGE_FORMAT_NV21 // or IMAGE_FORMAT_YV12
)
// Or:
val image = InputImage.fromByteArray(
        byteArray,
        /* image width */ 480,
        /* image height */ 360,
        rotationDegrees,
        InputImage.IMAGE_FORMAT_NV21 // or IMAGE_FORMAT_YV12
)

Java

InputImage image = InputImage.fromByteBuffer(byteBuffer,
        /* image width */ 480,
        /* image height */ 360,
        rotationDegrees,
        InputImage.IMAGE_FORMAT_NV21 // or IMAGE_FORMAT_YV12
);
// Or:
InputImage image = InputImage.fromByteArray(
        byteArray,
        /* image width */480,
        /* image height */360,
        rotation,
        InputImage.IMAGE_FORMAT_NV21 // or IMAGE_FORMAT_YV12
);

使用 Bitmap

如需基于 Bitmap 对象创建 InputImage 对象,请进行以下声明:

Kotlin

val image = InputImage.fromBitmap(bitmap, 0)

Java

InputImage image = InputImage.fromBitmap(bitmap, rotationDegree);

图片由 Bitmap 对象以及旋转角度表示。

处理图片

将图片传递给 process 方法:

Kotlin

val result = detector.process(image)
        .addOnSuccessListener { result ->
            // Task completed successfully
            // …
        }
        .addOnFailureListener { e ->
            // Task failed with an exception
            // …
        }

Java

Task<List<FaceMesh>> result = detector.process(image)
        .addOnSuccessListener(
                new OnSuccessListener<List<FaceMesh>>() {
                    @Override
                    public void onSuccess(List<FaceMesh> result) {
                        // Task completed successfully
                        // …
                    }
                })
        .addOnFailureListener(
                new OnFailureListener() {
                    @Override
                    Public void onFailure(Exception e) {
                        // Task failed with an exception
                        // …
                    }
                });

获取有关检测到的人脸网格的相关信息

如果在图片中检测到任何人脸,系统会向成功监听器传递一组 FaceMesh 对象。每个 FaceMesh 都代表在图片中检测到的一张人脸。对于每个人脸网格,您可以获取它在输入图片中的边界坐标,以及您已配置人脸网格检测器所要查找的任何其他信息。

Kotlin

for (faceMesh in faceMeshs) {
    val bounds: Rect = faceMesh.boundingBox()

    // Gets all points
    val faceMeshpoints = faceMesh.allPoints
    for (faceMeshpoint in faceMeshpoints) {
      val index: Int = faceMeshpoints.index()
      val position = faceMeshpoint.position
    }

    // Gets triangle info
    val triangles: List<Triangle<FaceMeshPoint>> = faceMesh.allTriangles
    for (triangle in triangles) {
      // 3 Points connecting to each other and representing a triangle area.
      val connectedPoints = triangle.allPoints()
    }
}

Java

for (FaceMesh faceMesh : faceMeshs) {
    Rect bounds = faceMesh.getBoundingBox();

    // Gets all points
    List<FaceMeshPoint> faceMeshpoints = faceMesh.getAllPoints();
    for (FaceMeshPoint faceMeshpoint : faceMeshpoints) {
        int index = faceMeshpoints.getIndex();
        PointF3D position = faceMeshpoint.getPosition();
    }

    // Gets triangle info
    List<Triangle<FaceMeshPoint>> triangles = faceMesh.getAllTriangles();
    for (Triangle<FaceMeshPoint> triangle : triangles) {
        // 3 Points connecting to each other and representing a triangle area.
        List<FaceMeshPoint> connectedPoints = triangle.getAllPoints();
    }
}