MediaPipe 0.8 : Colab : MediaPipe 顔検出 (翻訳/解説)

翻訳 : (株)クラスキャット セールスインフォメーション
作成日時 : 03/06/2021 (0.8.3)

* 本ページは、MediaPipe の以下のドキュメントを翻訳した上で適宜、補足説明したものです：

• MediaPipe Face Detection Colab

* サンプルコードの動作確認はしておりますが、必要な場合には適宜、追加改変しています。
* ご自由にリンクを張って頂いてかまいませんが、sales-info@classcat.com までご一報いただけると嬉しいです。

MediaPipe 0.8 : Colab : MediaPipe 顔検出

Python の MediaPipe 顔検出ソリューションの使用サンプルです (http://solutions.mediapipe.dev/face_detection もまた見てください)。

!pip install mediapipe

顔を含む任意の画像を Colab にアップロードします。web: https://unsplash.com/photos/JyVcAIUAcPM と https://unsplash.com/photos/auTAb39ImXg から 2 つのサンプル画像を取ります。

from google.colab import files

uploaded = files.upload()

import cv2
from google.colab.patches import cv2_imshow

# Read images with OpenCV.
images = {name: cv2.imread(name) for name in uploaded.keys()}

# Preview the images.
for name, image in images.items():
  print(name)   
  cv2_imshow(image)

総ての MediaPipe ソリューション Python API サンプルは mp.solutions 下にあります。

MediaPipe Face Mesh ソリューションについては、mp_face_detection = mp.solutions.face_detection としてモジュールにアクセスできます。

初期化の間にパラメータ min_detection_confidence を変更しても良いです。パラメータについてのより多くの情報を得るためには help(mp_face_detection.FaceDetection) を実行します。

import mediapipe as mp
mp_face_detection = mp.solutions.face_detection

help(mp_face_detection.FaceDetection)

Help on class FaceDetection in module mediapipe.python.solutions.face_detection:

class FaceDetection(mediapipe.python.solution_base.SolutionBase)
 |  FaceDetection(min_detection_confidence=0.5)
 |  
 |  MediaPipe Face Detection.
 |  
 |  MediaPipe Face Detection processes an RGB image and returns a list of the
 |  detected face location data.
 |  
 |  Please refer to
 |  https://solutions.mediapipe.dev/face_detection#python-solution-api
 |  for usage examples.
 |  
 |  Method resolution order:
 |      FaceDetection
 |      mediapipe.python.solution_base.SolutionBase
 |      builtins.object
 |  
 |  Methods defined here:
 |  
 |  __init__(self, min_detection_confidence=0.5)
 |      Initializes a MediaPipe Face Detection object.
 |      
 |      Args:
 |        min_detection_confidence: Minimum confidence value ([0.0, 1.0]) for face
 |          detection to be considered successful. See details in
 |          https://solutions.mediapipe.dev/face_detection#min_detection_confidence.
 |  
 |  process(self, image: numpy.ndarray) -> 
 |      Processes an RGB image and returns a list of the detected face location data.
 |      
 |      Args:
 |        image: An RGB image represented as a numpy ndarray.
 |      
 |      Raises:
 |        RuntimeError: If the underlying graph throws any error.
 |        ValueError: If the input image is not three channel RGB.
 |      
 |      Returns:
 |        A NamedTuple object with a "detections" field that contains a list of the
 |        detected face location data.
 |  
 |  ----------------------------------------------------------------------
 |  Methods inherited from mediapipe.python.solution_base.SolutionBase:
 |  
 |  __enter__(self)
 |      A "with" statement support.
 |  
 |  __exit__(self, exc_type, exc_val, exc_tb)
 |      Closes all the input sources and the graph.
 |  
 |  close(self) -> None
 |      Closes all the input sources and the graph.
 |  
 |  ----------------------------------------------------------------------
 |  Data descriptors inherited from mediapipe.python.solution_base.SolutionBase:
 |  
 |  __dict__
 |      dictionary for instance variables (if defined)
 |  
 |  __weakref__
 |      list of weak references to the object (if defined)

# Prepare DrawingSpec for drawing the face landmarks later.
mp_drawing = mp.solutions.drawing_utils 
drawing_spec = mp_drawing.DrawingSpec(thickness=1, circle_radius=1)

with mp_face_detection.FaceDetection(
    min_detection_confidence=0.5) as face_detection:
  for name, image in images.items():
    # Convert the BGR image to RGB and process it with MediaPipe Face Detection.
    results = face_detection.process(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))

    # Draw face detections of each face.
    print(f'Face detections of {name}:')
    if not results.detections:
      continue
    annotated_image = image.copy()
    for detection in results.detections:
      # print('Nose tip:')
      # print(mp.python.solutions.face_detection.get_key_point(
      #     detection, mp_face_detection.FaceKeyPoints.NOSE_TIP))
      mp_drawing.draw_detection(annotated_image, detection)
      cv2_imshow(annotated_image)

以上