Anna Lucia's profile - activity

15 seconds to start up UMat OpenCV3.1 Hi, I tested on my computer: Intel i7-6700, the GPU is HD Graphics 530. When I us

The version I used is opencv3.0.0.

Hi, I have a confusion that, when the image sequence to be processed using opencl, the context of opencl will be initialized many times, the initialization may make efficiency lower. The question is that, is it possible to initialize the openCL context ONCE during the image sequence process?

Are there versions 2.XX available on Centos7? Because the projects are compeleted using C.

The radius of the kernel is 3 for both cases. And I've tested both CPU and GPU.

Theoretically, the sepFilter is faster than Filter. But, the results in my tests show that the Filter2D is about 4 times faster than the sepFilter2D. I tested the two functions in CPU,with the same kernel radius size.

Have you tested other filter algorithms, are they still slower down? In my project the GaussianBlur is slower in GPU than CUP, but the other algorithms are more faster in GPU. Maybe the GaussianBlur is suitbale computing in CPU because of its theory.

The data member in UMat is hidden and there is no constructor with user data. And some functions can not be carried on OpenCL. So some Mat should be remained.

Yeah, but both the possibility of errors are ruled out.And the problem is still existing.

Hi, The bilateralFilter is used in the video image processing, and after a period of time, a dump happens. That's memory Free or Allocator breakpoint. The call stack reports the location is in ocl_bilateralFilter_8u.

Hi, The bilateralFilter is used in the video image processing, and after a period of time, a dump happens. That's memory Free or Allocator breakpoint. The call stack reports the location is in ocl_bilateralFilter_8u.

Hi, The bilateralFilter is used in the video image processing, and after a period of time, a dump happens. That's memory Free or Allocator breakpoint. The call stack reports the location is in ocl_bilateralFilter_8u.

Hi, the question is that wher is the data in oclMat initialization function, in host or device?

Hi, I'm trying to use OpenCL to accelerate the imgproc module. The test results show that the time costed by data transfering is nearly the same as the computing time, and that's a lager number. I use funcs getUMat and copyTo to do the data transfer. The GPU platform in my computer is Intel HD Graphics 4600, and that's the only GPU device.So I want to konw how does the data transfer between the device and host in OpenCV? Is the transfer mode suitable for Intel Graphics?

Hi, I'm trying to compare the performance of facedetect computing with OpenCL and without. So I use the Mat and UMat for the two cases. But the UMat is more slow than Mat. The OpenCL runtime is OK on my computer. The question is how can I start the GPU device?

How about trying following code: - current_device.set(current_device.ptr()); BTW, you'd beeter set the environment variables to set up the OpenCL device.

Yes, I've set the flag as true. But the result is the same as false when I use the haarcascade to do the facedetect. Then I change the cascade file as lbpcascade, and get a double FPS.

I've done some tests refer to http://answers.opencv.org/question/58.... For the filtering and Soble cases, I have got some good results. But the results of the facedetect demo are still bad. The GPU has seemingly not started up, because the GPU load(tested by GPU-Z) is 1%~2%. And I have not check out the reason until now.