NVIDIA jetson编译opencv 源码 python版本

安装 jetson-stats 查看GPU的利用率

sudo apt-get install python3-pip

sudo -H pip3 install jetson-stats

运行 jtop 进行查看

opencv 编译python版本 编译命令

cmake -D CMAKE_INSTALL_PREFIX=/usr/local/opencv-4.6.0 -D CMAKE_BUILD_TYPE=RELEASE -D WITH_OPENGL=ON  -D BUILD_TIFF=ON -D WITH_TBB=ON -D WITH_V4L=ON -D WITH_GTK_2_X=ON  -D WITH_QT=ON  -D WITH_V4L=ON   -D WITH_TBB=ON -D WITH_GTK=ON -D ENABLE_FAST_MATH=ON   -D WITH_FFMPEG=ON  -D WITH_CUDA=OFF  -D WITH_CUDNN=OFF -D CUDA_FAST_MATH=ON -D WITH_CUBLAS=ON  -D WITH_NVCUVID=ON  -D BUILD_TESTS=OFF -D BUILD_PERF_TESTS=OFF -D BUILD_JAVA=OFF  -D INSTALL_C_EXAMPLES=OFF  -D BUILD_EXAMPLES=OFF -D WITH_EIGEN=OFF -D BUILD_opencv_python2=OFF -D BUILD_opencv_python3=ON -D PYTHON3_LIBRARY=/home/nvidia/anaconda3/lib -D PYTHON3_INCLUDE_DIR=/home/nvidia/anaconda3/include  -D PYTHON3_INCLUDE_DIR=/usr/include/python3.8 -D PYTHON3_INCLUDE_PATH=/usr/include/python3.8 -D HAVE_opencv_python3=ON  -D  PYTHON3_NUMPY_INCLUDE_DIRS=/home/nvidia/anaconda3/lib/python3.8/site-packages/numpy/core/include -D PYTHON3_PACKAGES_PATH=/home/nvidia/anaconda3/lib/python3.8/site-packages OPENCV_EXTRA_MODULES_PATH=/home/nvidia/library/opencv_contrib-4.6.0/modules   ..

Anaconda arm/aarch64 版本

arm/aarch64架构安装conda环境

Index of /anaconda/archive/ | 清华大学开源软件镜像站 | Tsinghua Open Source Mirror

jetpack 的 torch版本

PyTorch for Jetson - Announcements - NVIDIA Developer Forums

PyTorch for Jetson - Announcements - NVIDIA Developer Forums