上一个教程 : 高级 API: 文本检测模型和文本识别模型
下一个教程 : 转换 PyTorch 分类模型并使用 OpenCV Python 发布
原作者 | Chengrui Wang, Yuantao Feng |
---|---|
兼容性 | OpenCV >= 4.5.4 |
简介
本节将介绍用于人脸检测的 cv::FaceDetectorYN 类和用于人脸识别的 cv::FaceRecognizerSF 类。
模型
本模块需要两个预先训练好的模型(ONNX 格式):
数据库 | 准确率 | 阈值 (normL2) | 阈值 (余弦) |
---|---|---|---|
LFW | 99.60% | 1.128 | 0.363 |
CALFW | 93.95% | 1.149 | 0.340 |
CPLFW | 91.05% | 1.204 | 0.275 |
AgeDB-30 | 94.90% | 1.202 | 0.277 |
CFP-FP | 94.80% | 1.253 | 0.212 |
代码
C++
- 可下载代码: 点击此处
- 代码一览:
#include <opencv2/dnn.hpp>
#include <opencv2/imgproc.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/objdetect.hpp>
#include <iostream>
using namespace cv;
using namespace std;
static
void visualize(Mat& input, int frame, Mat& faces, double fps, int thickness = 2)
{
std::string fpsString = cv::format("FPS : %.2f", (float)fps);
if (frame >= 0)
cout << "Frame " << frame << ", ";
cout << "FPS: " << fpsString << endl;
for (int i = 0; i < faces.rows; i++)
{
// 打印结果
cout << "Face " << i
<< ", top-left coordinates: (" << faces.at<float>(i, 0) << ", " << faces.at<float>(i, 1) << "), "
<< "box width: " << faces.at<float>(i, 2) << ", box height: " << faces.at<float>(i, 3) << ", "
<< "score: " << cv::format("%.2f", faces.at<float>(i, 14))
<< endl;
// 绘制边界框
rectangle(input, Rect2i(int(faces.at<float>(i, 0)), int(faces.at<float>(i, 1)), int(faces.at<float>(i, 2)), int(faces.at<float>(i, 3))), Scalar(0, 255, 0), thickness);
// 绘制地标
circle(input, Point2i(int(faces.at<float>(i, 4)), int(faces.at<float>(i, 5))), 2, Scalar(255, 0, 0), thickness);
circle(input, Point2i(int(faces.at<float>(i, 6)), int(faces.at<float>(i, 7))), 2, Scalar(0, 0, 255), thickness);
circle(input, Point2i(int(faces.at<float>(i, 8)), int(faces.at<float>(i, 9))), 2, Scalar(0, 255, 0), thickness);
circle(input, Point2i(int(faces.at<float>(i, 10)), int(faces.at<float>(i, 11))), 2, Scalar(255, 0, 255), thickness);
circle(input, Point2i(int(faces.at<float>(i, 12)), int(faces.at<float>(i, 13))), 2, Scalar(0, 255, 255), thickness);
}
putText(input, fpsString, Point(0, 15), FONT_HERSHEY_SIMPLEX, 0.5, Scalar(0, 255, 0), 2);
}
int main(int argc, char** argv)
{
CommandLineParser parser(argc, argv,
"{help h | | Print this message}"
"{image1 i1 | | Path to the input image1. Omit for detecting through VideoCapture}"
"{image2 i2 | | Path to the input image2. When image1 and image2 parameters given then the program try to find a face on both images and runs face recognition algorithm}"
"{video v | 0 | Path to the input video}"
"{scale sc | 1.0 | Scale factor used to resize input video frames}"
"{fd_model fd | face_detection_yunet_2021dec.onnx| Path to the model. Download yunet.onnx in https://github.com/opencv/opencv_zoo/tree/master/models/face_detection_yunet}"
"{fr_model fr | face_recognition_sface_2021dec.onnx | Path to the face recognition model. Download the model at https://github.com/opencv/opencv_zoo/tree/master/models/face_recognition_sface}"
"{score_threshold | 0.9 | Filter out faces of score < score_threshold}"
"{nms_threshold | 0.3 | Suppress bounding boxes of iou >= nms_threshold}"
"{top_k | 5000 | Keep top_k bounding boxes before NMS}"
"{save s | false | Set true to save results. This flag is invalid when using camera}"
);
if (parser.has("help"))
{
parser.printMessage();
return 0;
}
String fd_modelPath = parser.get<String>("fd_model");
String fr_modelPath = parser.get<String>("fr_model");
float scoreThreshold = parser.get<float>("score_threshold");
float nmsThreshold = parser.get<float>("nms_threshold");
int topK = parser.get<int>("top_k");
bool save = parser.get<bool>("save");
float scale = parser.get<float>("scale");
double cosine_similar_thresh = 0.363;
double l2norm_similar_thresh = 1.128;
// 初始化 FaceDetectorYN
Ptr<FaceDetectorYN> detector = FaceDetectorYN::create(fd_modelPath, "", Size(320, 320), scoreThreshold, nmsThreshold, topK);
TickMeter tm;
// 如果输入是图像
if (parser.has("image1"))
{
String input1 = parser.get<String>("image1");
Mat image1 = imread(samples::findFile(input1));
if (image1.empty())
{
std::cerr << "Cannot read image: " << input1 << std::endl;
return 2;
}
int imageWidth = int(image1.cols * scale);
int imageHeight = int(image1.rows * scale);
resize(image1, image1, Size(imageWidth, imageHeight));
tm.start();
// 在推理之前设置输入大小
detector->setInputSize(image1.size());
Mat faces1;
detector->detect(image1, faces1);
if (faces1.rows < 1)
{
std::cerr << "Cannot find a face in " << input1 << std::endl;
return 1;
}
tm.stop();
// 在输入图像上绘制结果
visualize(image1, -1, faces1, tm.getFPS());
// 如果保存为 true,则保存结果
if (save)
{
cout << "Saving result.jpg...\n";
imwrite("result.jpg", image1);
}
// 可视化结果
imshow("image1", image1);
pollKey(); // 处理用户界面事件以显示内容
if (parser.has("image2"))
{
String input2 = parser.get<String>("image2");
Mat image2 = imread(samples::findFile(input2));
if (image2.empty())
{
std::cerr << "Cannot read image2: " << input2 << std::endl;
return 2;
}
tm.reset();
tm.start();
detector->setInputSize(image2.size());
Mat faces2;
detector->detect(image2, faces2);
if (faces2.rows < 1)
{
std::cerr << "Cannot find a face in " << input2 << std::endl;
return 1;
}
tm.stop();
visualize(image2, -1, faces2, tm.getFPS());
if (save)
{
cout << "Saving result2.jpg...\n";
imwrite("result2.jpg", image2);
}
imshow("image2", image2);
pollKey();
// 初始化 FaceRecognizerSF
Ptr<FaceRecognizerSF> faceRecognizer = FaceRecognizerSF::create(fr_modelPath, "");
// 通过检测到的第一个人脸对齐和裁剪人脸图像。
Mat aligned_face1, aligned_face2;
faceRecognizer->alignCrop(image1, faces1.row(0), aligned_face1);
faceRecognizer->alignCrop(image2, faces2.row(0), aligned_face2);
// 使用给定的对齐面进行特征提取
Mat feature1, feature2;
faceRecognizer->feature(aligned_face1, feature1);
feature1 = feature1.clone();
faceRecognizer->feature(aligned_face2, feature2);
feature2 = feature2.clone();
double cos_score = faceRecognizer->match(feature1, feature2, FaceRecognizerSF::DisType::FR_COSINE);
double L2_score = faceRecognizer->match(feature1, feature2, FaceRecognizerSF::DisType::FR_NORM_L2);
if (cos_score >= cosine_similar_thresh)
{
std::cout << "They have the same identity;";
}
else
{
std::cout << "They have different identities;";
}
std::cout << " Cosine Similarity: " << cos_score << ", threshold: " << cosine_similar_thresh << ". (higher value means higher similarity, max 1.0)\n";
if (L2_score <= l2norm_similar_thresh)
{
std::cout << "They have the same identity;";
}
else
{
std::cout << "They have different identities.";
}
std::cout << " NormL2 Distance: " << L2_score << ", threshold: " << l2norm_similar_thresh << ". (lower value means higher similarity, min 0.0)\n";
}
cout << "Press any key to exit..." << endl;
waitKey(0);
}
else
{
int frameWidth, frameHeight;
VideoCapture capture;
std::string video = parser.get<string>("video");
if (video.size() == 1 && isdigit(video[0]))
capture.open(parser.get<int>("video"));
else
capture.open(samples::findFileOrKeep(video)); // 保留 GStreamer 管道
if (capture.isOpened())
{
frameWidth = int(capture.get(CAP_PROP_FRAME_WIDTH) * scale);
frameHeight = int(capture.get(CAP_PROP_FRAME_HEIGHT) * scale);
cout << "Video " << video
<< ": width=" << frameWidth
<< ", height=" << frameHeight
<< endl;
}
else
{
cout << "Could not initialize video capturing: " << video << "\n";
return 1;
}
detector->setInputSize(Size(frameWidth, frameHeight));
cout << "Press 'SPACE' to save frame, any other key to exit..." << endl;
int nFrame = 0;
for (;;)
{
// 获取帧
Mat frame;
if (!capture.read(frame))
{
cerr << "Can't grab frame!Stop\n";
break;
}
resize(frame, frame, Size(frameWidth, frameHeight));
// 推理
Mat faces;
tm.start();
detector->detect(frame, faces);
tm.stop();
Mat result = frame.clone();
// 在输入图像上绘制结果
visualize(result, nFrame, faces, tm.getFPS());
// 可视化结果
imshow("Live", result);
int key = waitKey(1);
bool saveFrame = save;
if (key == ' ')
{
saveFrame = true;
key = 0; // 处理
}
if (saveFrame)
{
std::string frame_name = cv::format("frame_%05d.png", nFrame);
std::string result_name = cv::format("result_%05d.jpg", nFrame);
cout << "Saving '" << frame_name << "' and '" << result_name << "' ...\n";
imwrite(frame_name, frame);
imwrite(result_name, result);
}
++nFrame;
if (key > 0)
break;
}
cout << "Processed " << nFrame << " frames" << endl;
}
cout << "Done." << endl;
return 0;
}
Python
- 可下载代码: 点击此处
- 代码一览:
import argparse
import numpy as np
import cv2 as cv
def str2bool(v):
if v.lower() in ['on', 'yes', 'true', 'y', 't']:
return True
elif v.lower() in ['off', 'no', 'false', 'n', 'f']:
return False
else:
raise NotImplementedError
parser = argparse.ArgumentParser()
parser.add_argument('--image1', '-i1', type=str, help='Path to the input image1. Omit for detecting on default camera.')
parser.add_argument('--image2', '-i2', type=str, help='Path to the input image2. When image1 and image2 parameters given then the program try to find a face on both images and runs face recognition algorithm.')
parser.add_argument('--video', '-v', type=str, help='Path to the input video.')
parser.add_argument('--scale', '-sc', type=float, default=1.0, help='Scale factor used to resize input video frames.')
parser.add_argument('--face_detection_model', '-fd', type=str, default='face_detection_yunet_2021dec.onnx', help='Path to the face detection model. Download the model at https://github.com/opencv/opencv_zoo/tree/master/models/face_detection_yunet')
parser.add_argument('--face_recognition_model', '-fr', type=str, default='face_recognition_sface_2021dec.onnx', help='Path to the face recognition model. Download the model at https://github.com/opencv/opencv_zoo/tree/master/models/face_recognition_sface')
parser.add_argument('--score_threshold', type=float, default=0.9, help='Filtering out faces of score < score_threshold.')
parser.add_argument('--nms_threshold', type=float, default=0.3, help='Suppress bounding boxes of iou >= nms_threshold.')
parser.add_argument('--top_k', type=int, default=5000, help='Keep top_k bounding boxes before NMS.')
parser.add_argument('--save', '-s', type=str2bool, default=False, help='Set true to save results. This flag is invalid when using camera.')
args = parser.parse_args()
def visualize(input, faces, fps, thickness=2):
if faces[1] is not None:
for idx, face in enumerate(faces[1]):
print('Face {}, top-left coordinates: ({:.0f}, {:.0f}), box width: {:.0f}, box height {:.0f}, score: {:.2f}'.format(idx, face[0], face[1], face[2], face[3], face[-1]))
coords = face[:-1].astype(np.int32)
cv.rectangle(input, (coords[0], coords[1]), (coords[0]+coords[2], coords[1]+coords[3]), (0, 255, 0), thickness)
cv.circle(input, (coords[4], coords[5]), 2, (255, 0, 0), thickness)
cv.circle(input, (coords[6], coords[7]), 2, (0, 0, 255), thickness)
cv.circle(input, (coords[8], coords[9]), 2, (0, 255, 0), thickness)
cv.circle(input, (coords[10], coords[11]), 2, (255, 0, 255), thickness)
cv.circle(input, (coords[12], coords[13]), 2, (0, 255, 255), thickness)
cv.putText(input, 'FPS: {:.2f}'.format(fps), (1, 16), cv.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)
if __name__ == '__main__':
detector = cv.FaceDetectorYN.create(
args.face_detection_model,
"",
(320, 320),
args.score_threshold,
args.nms_threshold,
args.top_k
)
tm = cv.TickMeter()
# 如果输入是图像
if args.image1 is not None:
img1 = cv.imread(cv.samples.findFile(args.image1))
img1Width = int(img1.shape[1]*args.scale)
img1Height = int(img1.shape[0]*args.scale)
img1 = cv.resize(img1, (img1Width, img1Height))
tm.start()
detector.setInputSize((img1Width, img1Height))
faces1 = detector.detect(img1)
tm.stop()
assert faces1[1] is not None, 'Cannot find a face in {}'.format(args.image1)
# 在输入图像上绘制结果
visualize(img1, faces1, tm.getFPS())
# 如果保存为 true,则保存结果
if args.save:
print('Results saved to result.jpg\n')
cv.imwrite('result.jpg', img1)
# 在新窗口中显示结果
cv.imshow("image1", img1)
if args.image2 is not None:
img2 = cv.imread(cv.samples.findFile(args.image2))
tm.reset()
tm.start()
detector.setInputSize((img2.shape[1], img2.shape[0]))
faces2 = detector.detect(img2)
tm.stop()
assert faces2[1] is not None, 'Cannot find a face in {}'.format(args.image2)
visualize(img2, faces2, tm.getFPS())
cv.imshow("image2", img2)
recognizer = cv.FaceRecognizerSF.create(
args.face_recognition_model,"")
face1_align = recognizer.alignCrop(img1, faces1[1][0])
face2_align = recognizer.alignCrop(img2, faces2[1][0])
# 提取特征
face1_feature = recognizer.feature(face1_align)
face2_feature = recognizer.feature(face2_align)
cosine_similarity_threshold = 0.363
l2_similarity_threshold = 1.128
cosine_score = recognizer.match(face1_feature, face2_feature, cv.FaceRecognizerSF_FR_COSINE)
l2_score = recognizer.match(face1_feature, face2_feature, cv.FaceRecognizerSF_FR_NORM_L2)
msg = 'different identities'
if cosine_score >= cosine_similarity_threshold:
msg = 'the same identity'
print('They have {}. Cosine Similarity: {}, threshold: {} (higher value means higher similarity, max 1.0).'.format(msg, cosine_score, cosine_similarity_threshold))
msg = 'different identities'
if l2_score <= l2_similarity_threshold:
msg = 'the same identity'
print('They have {}. NormL2 Distance: {}, threshold: {} (lower value means higher similarity, min 0.0).'.format(msg, l2_score, l2_similarity_threshold))
cv.waitKey(0)
else: # 省略输入,调用默认摄像机
if args.video is not None:
deviceId = args.video
else:
deviceId = 0
cap = cv.VideoCapture(deviceId)
frameWidth = int(cap.get(cv.CAP_PROP_FRAME_WIDTH)*args.scale)
frameHeight = int(cap.get(cv.CAP_PROP_FRAME_HEIGHT)*args.scale)
detector.setInputSize([frameWidth, frameHeight])
while cv.waitKey(1) < 0:
hasFrame, frame = cap.read()
if not hasFrame:
print('No frames grabbed!')
break
frame = cv.resize(frame, (frameWidth, frameHeight))
# 推断
tm.start()
faces = detector.detect(frame) # faces 是一个元组
tm.stop()
# 在输入图像上绘制结果
visualize(frame, faces, tm.getFPS())
# 可视化结果
cv.imshow('Live', frame)
cv.destroyAllWindows()
说明
C++
// Initialize FaceDetectorYN
Ptr<FaceDetectorYN> detector = FaceDetectorYN::create(fd_modelPath, "", Size(320, 320), scoreThreshold, nmsThreshold, topK);
// Set input size before inference
detector->setInputSize(image1.size());
Mat faces1;
detector->detect(image1, faces1);
if (faces1.rows < 1)
{
std::cerr << "Cannot find a face in " << input1 << std::endl;
return 1;
}
Python
detector = cv.FaceDetectorYN.create(
args.face_detection_model,
"",
(320, 320),
args.score_threshold,
args.nms_threshold,
args.top_k
)
# Set input size before inference
detector.setInputSize((img1Width, img1Height))
faces1 = detector.detect(img1)
检测输出的人脸是一个 CV_32F 类型的二维数组,其行是检测到的人脸实例,列是人脸的位置和 5 个人脸地标。每行的格式如下
x1,y1,w,h,x_re,y_re,x_le,y_le,x_nt,y_nt,x_rcm,y_rcm,x_lcm,y_lcm
其中,x1、y1、w、h
分别为人脸边界框的左上角坐标、宽度和高度,{x, y}_{re, le, nt, rcm, lcm}
分别代表右眼、左眼、鼻尖、右嘴角和左嘴角的坐标。
人脸识别
人脸检测后,运行以下代码从面部图像中提取人脸特征。
C++
// Initialize FaceRecognizerSF
Ptr<FaceRecognizerSF> faceRecognizer = FaceRecognizerSF::create(fr_modelPath, "");
// Aligning and cropping facial image through the first face of faces detected.
Mat aligned_face1, aligned_face2;
faceRecognizer->alignCrop(image1, faces1.row(0), aligned_face1);
faceRecognizer->alignCrop(image2, faces2.row(0), aligned_face2);
// Run feature extraction with given aligned_face
Mat feature1, feature2;
faceRecognizer->feature(aligned_face1, feature1);
feature1 = feature1.clone();
faceRecognizer->feature(aligned_face2, feature2);
feature2 = feature2.clone();
Python
recognizer = cv.FaceRecognizerSF.create(
args.face_recognition_model,"")
# Align faces
face1_align = recognizer.alignCrop(img1, faces1[1][0])
face2_align = recognizer.alignCrop(img2, faces2[1][0])
# Extract features
face1_feature = recognizer.feature(face1_align)
face2_feature = recognizer.feature(face2_align)
获得两张人脸图像的人脸特征 feature1 和 feature2 后,运行下面的代码来计算两张人脸的身份差异。
C++
double cos_score = faceRecognizer->match(feature1, feature2, FaceRecognizerSF::DisType::FR_COSINE);
double L2_score = faceRecognizer->match(feature1, feature2, FaceRecognizerSF::DisType::FR_NORM_L2);
Python
cosine_score = recognizer.match(face1_feature, face2_feature, cv.FaceRecognizerSF_FR_COSINE)
l2_score = recognizer.match(face1_feature, face2_feature, cv.FaceRecognizerSF_FR_NORM_L2)
例如,如果余弦距离大于或等于 0.363,或 normL2 距离小于或等于 1.128,则两张脸具有相同的身份。
参考文献
- https://github.com/ShiqiYu/libfacedetection
- https://github.com/ShiqiYu/libfacedetection.train
- https://github.com/zhongyy/SFace