本文主要是介绍python dlib 面部特征点检测，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！
运行环境macos m2芯片 Python 3.11.7，python3.9都能通过，windows系统应该也是一样的效果
import dlib
import cv2
import matplotlib.pyplot as plt# Load the image
image_path = 'path_to_your_image.jpg' # Replace with the path to your image
image = cv2.imread(image_path)# Convert the image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)# Initialize the dlib face detector
detector = dlib.get_frontal_face_detector()# Initialize the shape predictor
predictor_path = 'shape_predictor_68_face_landmarks.dat' # Replace with the path to your .dat file
predictor = dlib.shape_predictor(predictor_path)# Detect faces in the image
faces = detector(gray)# Loop over each detected face
for face in faces:# Get the landmarkslandmarks = predictor(gray, face)# Loop over each landmark pointfor n in range(0, 68):x = landmarks.part(n).xy = landmarks.part(n).ycv2.circle(image, (x, y), 2, (0, 255, 0), -1)# Convert the image back to RGB for displaying with matplotlib
image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)# Display the image with detected landmarks
plt.imshow(image_rgb)
plt.axis('off')
plt.show()
dlib 的亮点之一是其高精度的人脸检测和面部特征点检测功能，包含了用于检测 68 个面部特征点的预训练模型。
shape_predictor_68_face_landmarks.dat下载地址
链接: https://pan.baidu.com/s/1MhwLavLU2uWJnNo13gTLjQ 提取码: 5555
--来自百度网盘超级会员v9的分享
这篇关于python dlib 面部特征点检测的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！
