Compile OpenCV with CUDA support on OSX

Question

I'm trying to compile opencv 3.4 with CUDA support on OSX. CUDA is installed and works fine, but make gives me an error. I issue the following options with cmake:

cmake -DBUILD_TIFF=ON -DBUILD_opencv_java=OFF \
-DWITH_CUDA=ON -DWITH_OPENGL=ON -DWITH_OPENCL=ON \
-DWITH_IPP=ON -DWITH_TBB=ON -DWITH_EIGEN=ON \
-DWITH_V4L=ON -DWITH_VTK=OFF -DBUILD_TESTS=OFF \
-DBUILD_PERF_TESTS=OFF -DCMAKE_BUILD_TYPE=RELEASE \
-DPYTHON3_EXECUTABLE=~/Projects/rtmp-test/bin/python \
-DPYTHON3_PACKAGES_PATH=~/Projects/rtmp-test/lib/python3.8/site-packages/ \
-DPYTHON_NUMPY_INCLUDE_DIRS=~/Projects/rtmp-test/lib/python3.8/site-packages/numpy/core/include/numpy \
-DOPENCV_EXTRA_MODULES_PATH=~/Projects/opencv-cuda/opencv_contrib/modules ..

But when I run make, I get the following error:

[ 44%] Linking CXX static library ../lib/liblibprotobuf.a
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/ranlib: 
file: ../lib/liblibprotobuf.a(atomicops_internals_x86_gcc.cc.o) has no symbols

I read here that it could be because nvcc doesn't support clang 8.1.0 and I'm on Xcode 9.2. Anyone know how this can be resolved?

Any thought appreciated.