vps's profile - activity

Hi All, As shown in the picture. I have two end point of line. I want to draw the rectangle by considering these points are the middle points of the two sides. It will be great, if anyone suggest me the easy way.

Thank you.

It is working when I have installed a proper fourcc codec.

Unable to save video file from camera Dear all, I am trying to save the greyscale video from the camera but the output v

Yeah, I kept the maximum height and width of the image and it is working.

Capture the frame from iDS uEye camera. Dear all, I am using the USB 3.0 iDS uEye UI-3241LE-MGL model. I am trying to ge

Capture the frame from iDS uEye camera. Dear all, I am using the USB 3.0 iDS uEye UI-3241LE-MGL model. I am trying to ge

Thank you. I am able to calculate eigenvecors by PCA.

Thanks for the suggestion @LBerger, As suggested by @berak, I have used PCA method to calculate the eigenvectors.

Calculation of eigenvalues and eigenvectors for contour Hi, I am trying to calculate eigenvectors but I am getting huge

Calculate the disparity value for specific point Hi All, I have left and right rectified image. Suppose, I have p(xl, yl

Calculate the disparity value for only 1000 pixels Hi All, I have left and right rectified image. I am trying to calcula

Calculate the disparity value for 1000 points Hi All, I have left and right rectified image. I am trying to calculate di

Calculate the disparity value for 1000 points Hi All, I have left and right rectified image. I am trying to calculate di

@sturkmen It is not too much. For CPU version requires 75 msec and GPU HOG takes 50msec.

@berak..Nowadays, CUDA is very important to speed the process. Sure, if I achieve something important, I will happy to c

@berak..Thnak you for the links. But maybe the Cuda SVM version will be updated in the next OpenCV version release. :)

@berak..yeah, I have only this and hog GPU code. But maybe the Cuda SVM version will be updated in the next OpenCV versi

@berak..Thanks for the reply. You are correct :) Actually, I want to use the support vector file which I have trained by

Can you add your code in question? I have tried the same code with the same image. It is working for me.

I thought you are using the exact code with the same image. can you add your code in question? cause I have tried same c

Try this 1) Add HappyFish.jpg image inside the folder. 2) Then change.. String imageName( "HappyFish.jpg" ); or you ca

1) Add HappyFish.jpg image inside the folder. 2) Then change.. String imageName( "HappyFish.jpg" ); or you can give the

Hi, There is no HappyFish.jpg image inside your folder. Can you verify it?

Hi @berak, I have a few questions. 1) SVM is used in GPU version of HOG. Can we use of same detection method here? Line

SVM GPU Version Hi, I am using SVM. I have trained SVM and stored matrix in .xml file. I want to predict the label for e

SVM GPU Version Hi, I am using SVM. I have trained SVM and stored matrix in .xml file. I want to predict the label for e

OpenCV SVM on CUDA Hi, I am using SVM. I have trained SVM and stored matrix in .xml file. I want to predict the label fo

OpenCV SVM on CUDA Hi, I am using SVM. I have trained SVM and stored matrix in .xml file. I want to predict the label fo

SVM GPU Version Hi, I am using SVM. I have trained SVM and stored matrix in .xml file. I want to predict the label for e

SVM GPU Version Hi, I am using SVM. I have trained SVM and stored matrix in .xml file. I want to predict the label for e

Few suggestions: You can try Hough Circles detection method. Another observation, the circles are present the edges of

of course :) Just in case, the conversion code cvtColor(img, frame, COLOR_BGR2BGRA);

Hi, I am Using Cuda hog from OpenCV but the program is crashing at totalRect.size(). does anyone have an idea about this?

int main(int argc, const char * argv[]) { VideoCapture cap("leftFront.avi"); if (!cap.isOpened()) return -1;

Mat img;
cv::Ptr<cv::cuda::HOG> gpu_hog = cv::cuda::HOG::create(Size(64, 128),Size(16, 16),Size(8, 8),Size(8, 8),9);
    gpu_hog -> setSVMDetector(HOGDescriptor::getDefaultPeopleDetector());
namedWindow("video capture", CV_WINDOW_AUTOSIZE);

while (true)
{
    cap >> img;
    if (!img.data)
        continue;
    Point pt;
    cv::cuda::GpuMat roiImg;
    roiImg.upload(img);
    vector<Rect> totalRect, selectedRect;
    vector<double> conf;
    gpu_hog -> detectMultiScale(roiImg, totalRect);
    size_t i, j;
  //cout << totalRect.end() << endl;
    for (i = 0; i<totalRect.size(); i++)
    {
        Rect r = totalRect[i];
        for (j = 0; j<totalRect.size(); j++)
            if (j != i && (r & totalRect[j]) == r)
                break;
        if (j == totalRect.size())
            selectedRect.push_back(r);
    }

    for (i = 0; i<selectedRect.size(); i++)
    {
        Rect r = selectedRect[i];
        r.x += cvRound(r.width*0.1);
        r.width = cvRound(r.width*0.8);
        r.y += cvRound(r.height*0.07);
        r.height = cvRound(r.height*0.8);
        rectangle(img, r.tl(), r.br(), Scalar(0, 255, 0), 3);
        //
        pt.x = r.tl().x + cvRound(r.width / 2);
        pt.y = r.tl().y + cvRound(r.height / 2);
        cout << pt.x << pt.y << endl;

        circle(img, Point(pt.x, pt.y), 5, Scalar(0, 255, 0), 5, 8, 0);
        //circle(roiImg, Point(r.tl().x, r.tl().y), 10, Scalar(255, 0, 0), 8, 8, 0);
        //
    }

    imshow("Person Detection", img);
    if (waitKey(33) >= 0)
        break;
}
return 0;

}

Hi @berak, I have solved the problem. I need to convert an image from BGR to BGRA for GPU HOG. Can you please close this

Hi @berak, I have solved the problem. I need to convert an image from BGR to BGRA for GPU hOG. Can you please close this

I think that processing on 16-bit image is not that difficult. What do you want to do exactly?

It is working without cuda.

Problem with cuda hog Hi, I am Using Cuda hog from OpenCV but the program is crashing at totalRect.size(). does anyone h