diff --git a/modules/gpuarithm/include/opencv2/gpuarithm.hpp b/modules/gpuarithm/include/opencv2/gpuarithm.hpp
index 4272e15468..5c51186716 100644
--- a/modules/gpuarithm/include/opencv2/gpuarithm.hpp
+++ b/modules/gpuarithm/include/opencv2/gpuarithm.hpp
@@ -205,6 +205,10 @@ inline void LUT(InputArray src, InputArray lut, OutputArray dst, Stream& stream)
     createLookUpTable(lut)->transform(src, dst, stream);
 }
 
+//! copies 2D array to a larger destination array and pads borders with user-specifiable constant
+CV_EXPORTS void copyMakeBorder(InputArray src, OutputArray dst, int top, int bottom, int left, int right, int borderType,
+                               Scalar value = Scalar(), Stream& stream = Stream::Null());
+
 //! implements generalized matrix product algorithm GEMM from BLAS
 CV_EXPORTS void gemm(const GpuMat& src1, const GpuMat& src2, double alpha,
     const GpuMat& src3, double beta, GpuMat& dst, int flags = 0, Stream& stream = Stream::Null());
@@ -273,10 +277,6 @@ CV_EXPORTS void meanStdDev(const GpuMat& mtx, Scalar& mean, Scalar& stddev, GpuM
 //! output will have CV_32FC1 type
 CV_EXPORTS void rectStdDev(const GpuMat& src, const GpuMat& sqr, GpuMat& dst, const Rect& rect, Stream& stream = Stream::Null());
 
-//! copies 2D array to a larger destination array and pads borders with user-specifiable constant
-CV_EXPORTS void copyMakeBorder(const GpuMat& src, GpuMat& dst, int top, int bottom, int left, int right, int borderType,
-                               const Scalar& value = Scalar(), Stream& stream = Stream::Null());
-
 //! computes the integral image
 //! sum will have CV_32S type, but will contain unsigned int values
 //! supports only CV_8UC1 source type
diff --git a/modules/gpuarithm/src/core.cpp b/modules/gpuarithm/src/core.cpp
index 5dc1d4a5e5..22887796ab 100644
--- a/modules/gpuarithm/src/core.cpp
+++ b/modules/gpuarithm/src/core.cpp
@@ -59,7 +59,7 @@ void cv::gpu::flip(InputArray, OutputArray, int, Stream&) { throw_no_cuda(); }
 
 Ptr<LookUpTable> cv::gpu::createLookUpTable(InputArray) { throw_no_cuda(); return Ptr<LookUpTable>(); }
 
-void cv::gpu::copyMakeBorder(const GpuMat&, GpuMat&, int, int, int, int, int, const Scalar&, Stream&) { throw_no_cuda(); }
+void cv::gpu::copyMakeBorder(InputArray, OutputArray, int, int, int, int, int, Scalar, Stream&) { throw_no_cuda(); }
 
 #else /* !defined (HAVE_CUDA) */
 
@@ -529,14 +529,17 @@ typedef Npp32s __attribute__((__may_alias__)) Npp32s_a;
 typedef Npp32s Npp32s_a;
 #endif
 
-void cv::gpu::copyMakeBorder(const GpuMat& src, GpuMat& dst, int top, int bottom, int left, int right, int borderType, const Scalar& value, Stream& s)
+void cv::gpu::copyMakeBorder(InputArray _src, OutputArray _dst, int top, int bottom, int left, int right, int borderType, Scalar value, Stream& _stream)
 {
-    CV_Assert(src.depth() <= CV_32F && src.channels() <= 4);
-    CV_Assert(borderType == BORDER_REFLECT_101 || borderType == BORDER_REPLICATE || borderType == BORDER_CONSTANT || borderType == BORDER_REFLECT || borderType == BORDER_WRAP);
+    GpuMat src = _src.getGpuMat();
 
-    dst.create(src.rows + top + bottom, src.cols + left + right, src.type());
+    CV_Assert( src.depth() <= CV_32F && src.channels() <= 4 );
+    CV_Assert( borderType == BORDER_REFLECT_101 || borderType == BORDER_REPLICATE || borderType == BORDER_CONSTANT || borderType == BORDER_REFLECT || borderType == BORDER_WRAP );
 
-    cudaStream_t stream = StreamAccessor::getStream(s);
+    _dst.create(src.rows + top + bottom, src.cols + left + right, src.type());
+    GpuMat dst = _dst.getGpuMat();
+
+    cudaStream_t stream = StreamAccessor::getStream(_stream);
 
     if (borderType == BORDER_CONSTANT && (src.type() == CV_8UC1 || src.type() == CV_8UC4 || src.type() == CV_32SC1 || src.type() == CV_32FC1))
     {