Himanshu's profile - activity

You got any lead on this?

@berak @jamesnzt - I have tried that cout (fout) operation to write the MAT in the file. But it still saves it in a YAML format with first two lines as header and then a field called data which contains the values. I just want to write the values in a file. It seems OpenCV can only save it as YAML or XML format.

Hi,

FileStorage allows us to save XML or YML in compressed format too. The example in OpenCV documentation says '.gz' format. Has anyone tried saving it in '.zip' format?

Thanks.

Hi,

I am using FileStorage to write out Mat object (in my case it's a depthMap from Kinect). FileStorage either saves it as XML or YAML with headers. I tried using 'for loop' to save only the data, but its time consuming. Is there a way where I can save only the data of MAT object directly in a file without the headers?

Thanks.

Thanks! From where can I get ffmpeg libs for centos to install with opencv?

Hi, Hi,

Can anyone list steps for installing OpenCV on Linux? I have CentOS.

Thanks!

Hi,

I also have the same problem of setting the frame rate of my webcam using the CV_CAP_PROP_FPS.

Did you find any solution?

Thanks and regards Himanshu

Hi,

I am using M.t() transpose function, but the output is incorrect:

Input matrix M:

 1    -2     1     0     0     0     0     0
 0     1    -2     1     0     0     0     0
 0     0     1    -2     1     0     0     0
 0     0     0     1    -2     1     0     0
 0     0     0     0     1    -2     1     0
 0     0     0     0     0     1    -2     1

After transposing (M.t()), I am getting following output which is wrong:

 1    -2     0     1     0     0
 1     0    -2     1     1    -2
 0     0     0     0     1     0
 0     0     0     0     0     0
 0     0     0     0     0     0
 0     0     0     0     0     0
 0     0     0     0     0     0
 0     0     0     0     0     0

Can anyone tell me why is this happening?

Thanks.

Hi,

Has anyone tried configuring OpenCV(2.4.2) including libdc1394 on Windows??

Thanks and regards Himanshu

Thanks; It's a good solution!

Hi,

Is this the right way of reading pixel values from a frame of video?:

IplImage* bgr_frame = cvQueryFrame( capture );
int width, height, nchannels, step;
int i, j, r_ch, b_ch, g_ch;

width = bgr_frame->width;
height = bgr_frame->height;
nchannels = bgr_frame->nChannels;
step = bgr_frame->widthStep;
uchar *data = ( uchar* )bgr_frame->imageData;

for(i = 0 ; i < height ; i++) 
{
  for( j = 0 ; j < width ; j++)
  {
    b_ch = ((uchar *)(bgr_frame->imageData + i*bgr_frame->widthStep))[j*bgr_frame->nChannels + 0];
    g_ch = ((uchar *)(bgr_frame->imageData + i*bgr_frame->widthStep))[j*bgr_frame->nChannels + 1];
    r_ch = ((uchar *)(bgr_frame->imageData + i*bgr_frame->widthStep))[j*bgr_frame->nChannels + 2];       
  }
}

Thanks and regards Himanshu

Hi,

I used a Logitech web-camera for live capture and it worked.

How to use a User RGB Camera (Olympus, Sony etc) with OpenCV on windows?? cvCaptureFromCAM doesn't recognize that camera!! Is libdc1394 library required??

Thanks and regards Himanshu

t is expressed in microseconds! so used 1000000/t; but the problem now i am facing is the FPS is varying!

Hi,

Martin, thanks for your reply.

I did what you suggested and i am still getting approx. 0 FPS.

Thanks.

Hi,

I know to find frame rate from a video, we can use the capture property and ID:CV_CAP_PROP_FPS.

But how to find the frame rate of a live capture? the capture property (CV_CAP_PROP_FPS) doesn't work for live capture.

Thanks.