Ask Your Question

grtwa's profile - activity

2017-04-13 10:37:00 -0500 received badge  Notable Question (source)
2016-01-02 07:10:33 -0500 received badge  Popular Question (source)
2014-05-10 19:20:29 -0500 commented question TBB cvtColor memory leaks

Thank you. I rebuild OpenCV 2.4.9 with TBB 4.2, but the memory leaks still exist.

2014-05-10 08:35:37 -0500 asked a question TBB cvtColor memory leaks

I build OpenCV 2.4.9 with TBB 4.1, and call cvtColor in the following two pieces of code. Both of them report memory leaks. How can I avoid the memory leaks? Many thanks.

void main()

{ _CrtSetDbgFlag ( _CRTDBG_ALLOC_MEM_DF | _CRTDBG_LEAK_CHECK_DF );

cv::Mat colorImage(320,480,CV_8UC3);
cv::Mat grayImage;
cv::cvtColor(colorImage,grayImage,CV_BGR2GRAY);

}

void main()

{ _CrtSetDbgFlag ( _CRTDBG_ALLOC_MEM_DF | _CRTDBG_LEAK_CHECK_DF );

cv::Mat colorImage(320,480,CV_8UC3);
cv::Mat grayImage(320,480,CV_8UC1);
cv::cvtColor(colorImage,grayImage,CV_BGR2GRAY);

colorImage.release();
grayImage.release();

}

2013-08-13 02:27:05 -0500 commented answer linear least-squares for large-scale dense matrix

Thanks, I need deal with some degenerate case. Before using SVD, I intended to use QR. But DECOMP_QR seems to equate with DECOMP_SVD in OpenCV2.4.5.

2013-08-12 20:00:58 -0500 received badge  Supporter (source)
2013-08-12 02:29:35 -0500 received badge  Self-Learner (source)
2013-08-11 23:36:52 -0500 answered a question linear least-squares for large-scale dense matrix

I have solved this problem by convert the compiling platform to x64. On x86 platform, only 2G physical memory can be used.

2013-08-11 23:33:22 -0500 commented answer I think the Gaussian weighting of SIFT descriptor is wrong

have done, thanks :)

2013-08-11 23:25:34 -0500 received badge  Scholar (source)
2013-08-11 23:25:34 -0500 received badge  Enlightened (source)
2013-08-09 22:46:01 -0500 received badge  Good Answer (source)
2013-08-09 05:32:34 -0500 received badge  Nice Answer (source)
2013-08-09 05:02:16 -0500 received badge  Teacher (source)
2013-08-09 04:50:08 -0500 received badge  Self-Learner (source)
2013-08-09 04:43:28 -0500 commented answer I think the Gaussian weighting of SIFT descriptor is wrong

Sorry, I have not 50 points to accept my own answer.

2013-08-09 04:37:46 -0500 answered a question I think the Gaussian weighting of SIFT descriptor is wrong

OpenCV is correct. Similar with vlfeat,"c_rot" and "r_rot" have been normalized by "hist_width", so the sigma is "d/2".

2013-08-09 04:35:41 -0500 commented question linear least-squares for large-scale dense matrix

@StevenPuttemans, "New users must wait 2 days before answering their own question." I will post an answer later. Thank you for reminding.

2013-08-09 04:31:07 -0500 commented question linear least-squares for large-scale dense matrix

@Moster, I did not try because I think automatic variables are created on the heap. Any way, problem is gone :)

2013-08-09 02:42:26 -0500 answered a question Best points for opticalFlow

For the LK, strong corner are the best points. I think FAST is good enough for tracking.

2013-08-09 01:28:51 -0500 commented question linear least-squares for large-scale dense matrix

Thank you. I have solved this problem by convert the compiling platform to x64. On x86 platform, only 2G physical memory can be used. It's my fault, not OpenCV :)

2013-08-09 01:02:16 -0500 received badge  Critic (source)
2013-08-08 04:48:58 -0500 asked a question linear least-squares for large-scale dense matrix

I use the function "cv::solve(src1, src2, dst, DECOMP_SVD)" to solve a least-squares problem (i.e., src1 * dst = src2). The size of src1 is 29030 * 7809 and the size of src2 is 29030 * 122. Both types of src1 and src2 are CV_32FC1. OpenCV crashes at the 1370 line of "lapack.cpp" and the code of this line is "buffer.allocate(bufsize)". "buffer" is a "AutoBuffer< uchar >" and "bufsize" is 1151016404. Is this problem caused by out-of-memory? My physical memory is 8G and operating system is 64bit Win7.

Mat A(29030,7809,CV_32FC1);
Mat b(29030,122,CV_32FC1);
for(i=0;i<29030;i++)
{
    for(j=0;j<7809;j++)
        A.at<float>(i,j)=float(i*7809+j);

    for(j=0;j<122;j++)
        b.at<float>(i,j)=float(i*122+j);
}

Mat x;
solve(A,b,x,DECOMP_SVD);
2013-08-01 22:53:59 -0500 commented question full screen video frame in iOS

Which app you downloaded? Can you modify its source code?

2013-08-01 10:02:46 -0500 answered a question classifier for eyeborow. help !

As far as I know, there is no open source code. Maybe you can classify them by shape and I think it is not a good idea to classify by key points and descriptors.

2013-08-01 09:45:20 -0500 commented question I think the Gaussian weighting of SIFT descriptor is wrong

I see. Similar with vlfeat,"c_rot" and "r_rot" have been normalized by "hist_width", so the sigma is "d/2". Thanks for Guanta and StevenPuttemans' help.

2013-08-01 05:22:51 -0500 received badge  Student (source)
2013-08-01 03:01:37 -0500 commented question I think the Gaussian weighting of SIFT descriptor is wrong

Sorry, it is not exact. I think 32x32 means (2xwidth+1)x(2xwidth+1), so width is 15.5 and Sigma is 7.75.

2013-07-31 21:37:47 -0500 received badge  Editor (source)
2013-07-31 21:36:49 -0500 asked a question I think the Gaussian weighting of SIFT descriptor is wrong

In David G. Lowe's paper, the Sigma of Gaussian weighting function is one half the width of the descriptor window. If the descriptor window is 32x32 pixels with 4x4 subregions, I think the Sigma should be 8 rather than 2. But in sift.cpp, the code is "float exp_scale = -1.f/(d * d * 0.5f)", which means the Sigma is 2. Is there anything wrong?

2013-02-07 00:13:46 -0500 answered a question Trouble linkig opencv_world from latest source

I also suffered this problem. Everything went well when compiling without CUDA,but I need to use CUDA. Any help is greatly appreciated.