如何使用 Dlib 的多目标检测器？答案

【问题标题】：How to use multiple object detector of Dlib?如何使用 Dlib 的多目标检测器？
【发布时间】：2018-06-22 16:02:49
【问题描述】：

我已经使用 Dlib 训练了两个人脸检测器。所以，我正在尝试使用那个（两个）人脸检测器来增加检测人脸。

object_detector1.svm
object_detector2.svm

但是，我不明白，如何使用它。请有人帮助我。

我看到this堆栈溢出问题并尝试使用它，但我得到一个错误。

我的努力：

#include <dlib/image_processing/frontal_face_detector.h>
#include <dlib/gui_widgets.h>
#include <dlib/image_io.h>
#include <iostream>

using namespace dlib;
using namespace std;

int main(int argc, char** argv)
{  
    try
    {
        if (argc == 1)
        {
            cout << "Give some image files as arguments to this program." << endl;
            return 0;
        }

        typedef scan_fhog_pyramid<pyramid_down<6> > image_scanner_type; 
        image_scanner_type scanner;

        object_detector<image_scanner_type> face_detector(2);
        dlib::deserialize(detectors[0], "object_detector1.svm");
        deserialize(face_detector[0],"object_detector2.svm");
        deserialize(face_detector[1],"face_detector1.svm");
        image_window win;

        // Loop over all the images provided on the command line.
        for (int i = 1; i < argc; ++i)
        {
            cout << "processing image " << argv[i] << endl;
            array2d<unsigned char> img;
            load_image(img, argv[i]);

            pyramid_up(img);

            std::vector<rectangle> dets = face_detector(img);

            cout << "Number of faces detected: " << dets.size() << endl;

            win.clear_overlay();
            win.set_image(img);
            win.add_overlay(dets, rgb_pixel(255,0,0));

            cout << "Hit enter to process the next image..." << endl;
            cin.get();
        }
    }
    catch (exception& e)
    {
        cout << "\nexception thrown!" << endl;
        cout << e.what() << endl;
    }
}

提前致谢。

【问题讨论】：

标签： c++ c++11 training-data dlib

【解决方案1】：

您可以将以下代码用于多个分类器。

object_detector<image_scanner_type> face_detector, face_detector2;
deserialize("object_detector1.svm")>>face_detector;
deserialize("object_detector2.svm")>>face_detector2;

std::vector<object_detector<image_scanner_type> > my_detectors;
my_detectors.push_back(face_detector);
my_detectors.push_back(face_detector2);

image_window win;

// Loop over all the images provided on the command line.
for (int i = 1; i < argc; ++i)
{
    cout << "processing image " << argv[i] << endl;
    array2d<unsigned char> img;
    load_image(img, argv[i]);

    pyramid_up(img);

    std::vector<rectangle> dets = evaluate_detectors(my_detectors, img);

    cout << "Number of faces detected: " << dets.size() << endl;

    win.clear_overlay();
    win.set_image(img);
    win.add_overlay(dets, rgb_pixel(255,0,0));

    cout << "Hit enter to process the next image..." << endl;
    cin.get();
}

【讨论】：