Dlib提取人脸特征点

转自https://blog.csdn.net/zmdsjtu/article/details/53454071?ticket=ST-3153-47Plcph2ez6yRnbTZqXP-passport.csdn.net

主要在官网给的Demo基础之上用Opencv把特征点描绘出来了。

很早之前写过一篇配置Dlib环境的博客，现在来稍微梳理下提取特征点的使用方法。

上一篇配置环境博客地址：http://blog.csdn.net/zmdsjtu/article/details/52422847

惯例先放效果图吧：

动图如下：

接着就是简单粗暴的代码：

1. //@zmdsjtu@163.com
2. //2016-12-4
3. //http://blog.csdn.net/zmdsjtu/article/details/53454071
4. #include <dlib/opencv.h>
5. #include <opencv2/opencv.hpp>
6. #include <dlib/image_processing/frontal_face_detector.h>
7. #include <dlib/image_processing/render_face_detections.h>
8. #include <dlib/image_processing.h>
9. #include <dlib/gui_widgets.h>
11. using namespace dlib;
12. using namespace std;
14. int main()
15. {
16. try
17. {
18. cv:: VideoCapture cap(0);
19. if (!cap.isOpened())
20. {
21. cerr << "Unable to connect to camera" << endl;
22. return 1;
23. }
25. //image_window win;
27. // Load face detection and pose estimation models.
28. frontal_face_detector detector = get_frontal_face_detector();
29. shape_predictor pose_model;
30. deserialize( "shape_predictor_68_face_landmarks.dat") >> pose_model;
32. // Grab and process frames until the main window is closed by the user.
33. while (cv::waitKey( 30) != 27)
34. {
35. // Grab a frame
36. cv::Mat temp;
37. cap >> temp;
39. cv_image<bgr_pixel> cimg(temp);
40. // Detect faces
41. std:: vector<rectangle> faces = detector(cimg);
42. // Find the pose of each face.
43. std:: vector<full_object_detection> shapes;
44. for ( unsigned long i = 0; i < faces.size(); ++i)
45. shapes.push_back(pose_model(cimg, faces[i]));
47. if (!shapes.empty()) {
48. for ( int i = 0; i < 68; i++) {
49. circle(temp, cvPoint(shapes[ 0].part(i).x(), shapes[ 0].part(i).y()), 3, cv::Scalar( 0, 0, 255), -1);
50. // shapes[0].part(i).x();//68个
51. }
52. }
53. //Display it all on the screen
54. imshow( "Dlib特征点", temp);
56. }
57. }
58. catch (serialization_error& e)
59. {
60. cout << "You need dlib's default face landmarking model file to run this example." << endl;
61. cout << "You can get it from the following URL: " << endl;
62. cout << " http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2" << endl;
63. cout << endl << e.what() << endl;
64. }
65. catch (exception& e)
66. {
67. cout << e.what() << endl;
68. }
69. }

来看下上面那段代码，所有的需要的特征点都存储在Shapes里。仔细看看下面这行代码：

circle(temp, cvPoint(shapes[0].part(i).x(), shapes[0].part(i).y()), 3, cv::Scalar(0, 0, 255), -1);

可以看到shpes[0]代表的是第一个人（可以同时检测到很多个人），part(i)代表的是第i个特征点，x()和y()是访问特征点坐标的途径。

每个特征点的编号如下：

在上述画图的基础上加了如下一行代码：

putText(temp, to_string(i), cvPoint(shapes[0].part(i).x(), shapes[0].part(i).y()), CV_FONT_HERSHEY_PLAIN, 1, cv::Scalar(255, 0, 0),1,4);

效果图：

对照着上图，比如说想获取鼻尖的坐标，那么横坐标就是shapes[0].part[30].x()，其余的类似。

在这个的基础上就可以做很多有意思的事情啦，2333

最后祝大家开发愉快：）