OpenCV的predict（）与detectMultiScale（）

Question

I'm working on doing some Face, Gender, and Age detection using OpenCV. I have a bunch of images I use to train the models, Essentally I currently have the following:

Ptr<cv::face::FaceRecognizer> model = cv::face::LBPHFaceRecognizer::create(9, 9);
std::vector<int> labels;
std::vector<std::string> imageFileNames;

for (int currImageIndex = 0; currImageIndex < imageFileNames.size(); currImageIndex++)
{
    cv::Mat currMatrix;
    std::string currentFileName = imageFileNames[currImageIndex];
    std::string gender;
    int currID = -1;

    //Save the image and the corresponding ID to the list(s).
    currMatrix = imread(currentFileName , CV_LOAD_IMAGE_GRAYSCALE);
    if (currMatrix.data != NULL)
    {
        images.push_back(currMatrix);
        labels.push_back(currID);
    }
}

model->train(images, labels);
model->write("C:\\temp.xml");

Then using the temp.xml heuristic, I predict the geneder like so:

gendermodel->predict(currMat, predictedLabel, conf);

However, I came across this implementation using detectMultiScale() and a "Cascade Classifier" . What is the diffrence? Is there a performance advantage to using a Cascade Classifier vs the way I am currently doing it? Does detectMultiScale() work better then predict() ?

Answer 1

CascadeClassifier::detectMultiScale function is used for object detection . It returns a variable of type std::vector<cv::Rect> which contains bounding rectangles of detected objects.

FaceRecognizer::predict function is used for object classification . It returns the class label of input image and the confidence with which the object is predicted.