Cuda findMinMax missing?

Does anyone know where the cuda version of findMinMax is hiding? I have seen references to it online but no luck trying to build a program with 4.5.1.

Have you built with CUDA and included the opencv2/cudaarithm.hpp header?

First, thanks for your quick reply. Yes, I THINK I did, that is, I selected “build cuda stubs” and “opencv-dnn” in the CMake configuration. But, that header file does not exist. Did I fail to select it properly, or not download everything?
Thanks again!

I have never built with BUILD_CUDA_STUBS flag but I would assume it would work. When you say the header doesn’t exist do you mean in the source or in a build produced install folder? Are you on windows or linux?

Does your cmake output of modules to be built include the cuda modules

--   OpenCV modules:
--     To be built:                 aruco bgsegm bioinspired calib3d ccalib core cudaarithm cudabgsegm cudacodec cudafeatures2d cudafilters cudaimgproc cudalegacy cudaobjdetect cudaoptflow cudastereo cudawarping cudev datasets dnn dnn_objdetect dpm face features2d flann fuzzy hfs highgui img_hash imgcodecs imgproc line_descriptor ml objdetect optflow phase_unwrapping photo plot python2 python3 quality reg rgbd saliency shape stereo stitching structured_light superres surface_matching text tracking ts video videoio videostab world xfeatures2d ximgproc xobjdetect xphoto

I am on Windows 10. And I mean that I search the entire directory structure at the point of the download and that file does not exist.
Where is that CMake output? I am using VS 2019 and I see the CMake.txt but it does not have that line. I think you are on to something here.

Hey Carl, are you building with opencv contrib as this is required for building with CUDA? The header can be found

The CMake output will be shown at the cmd prompt if you call CMake from there or in the bottom window of the CMake GUI if you use that.

You may find the guide below useful for building OpenCV with CUDA on Windows.

Aha! I am not doing the contrib
Thanks much!

James, thank you so much for your help.

1 Like