moved GpuMat's operations implementation to core module
This commit is contained in:
@@ -139,10 +139,6 @@ private:
|
||||
int minorVersion_;
|
||||
};
|
||||
|
||||
//////////////////////////////// Error handling ////////////////////////
|
||||
|
||||
CV_EXPORTS void error(const char *error_string, const char *file, const int line, const char *func);
|
||||
|
||||
//////////////////////////////// CudaMem ////////////////////////////////
|
||||
// CudaMem is limited cv::Mat with page locked memory allocation.
|
||||
// Page locked memory is only needed for async and faster coping to GPU.
|
||||
|
Reference in New Issue
Block a user