-
Notifications
You must be signed in to change notification settings - Fork 61
Compile manual v1.1.0
taotianran edited this page Feb 7, 2023
·
2 revisions
For easy compilation, we provide docker images for different platforms, you can also compile on your local machine.
Go to install docker(optional): Docker Installation
# clone code
git clone git@github.com:PaddlePaddle/FlyCV.git
git checkout <branch name or tag name>
# set the ndk path, the NDK_PATH needs to be set to the NDK path on your local machine
export ANDROID_NDK=${NDK_PATH}
# compile
cd FlyCV
./scripts/android/build.sh
docker pull flycv/x86_64:android
docker attach
source /etc/ndk/r17c.env
./scripts/android/build.sh
# clone code
git clone git@github.com:PaddlePaddle/FlyCV.git
git checkout <branch name or tag name>
# compile
cd FlyCV
./scripts/armlinux/build.sh
# clone code
git clone git@github.com:PaddlePaddle/FlyCV.git
git checkout <branch name or tag name>
# compile
# if your computer is x86 architecture, execute the following command
cd FlyCV
./scripts/macos/build_x86.sh
# if your computer is arm architecture, execute the following command
./scripts/macos/build_arm.sh
# clone code
git clone git@github.com:PaddlePaddle/FlyCV.git
git checkout <branch name or tag name>
# compile
# generate solutions with the following command
./scripts/windows/build.sh
# open the solution to compile with Visual Studio
# clone code
git clone git@github.com:PaddlePaddle/FlyCV.git
git checkout <branch name or tag name>
# compile
./scripts/linux/build.sh
compile options | default | description |
---|---|---|
📃 FLYCV_NAMESPACE | flycv | you can change to any customized namespace to avoid library conflicts |
compile options | default | description |
---|---|---|
🗂️ BUILD_FCV_CORE | ON | core functions cannot be turned off |
🗂️ BUILD_FCV_IMG_TRANSFORM | ON | image transform functions |
📃 WITH_FCV_COLOR_CONVERT | ON | supports frequently used color space conversion functions |
📃 WITH_FCV_RESIZE | ON | image resize |
📃 WITH_FCV_WARP_AFFINE | ON | image affine transformation |
📃 WITH_FCV_WARP_PERSPECTIVE | ON | image perspective transformation |
📃 WITH_FCV_ROTATION | ON | image rotation |
📃 WITH_FCV_FLIP | ON | image flipped along the axes |
📃 WITH_FCV_SUBTRACT | ON | image subtraction |
📃 WITH_FCV_COPY_MAKE_BORDER | ON | image border padding |
📃 WITH_FCV_CROP | ON | crop image |
📃 WITH_FCV_REMAP | ON | image remap |
📃 WITH_FCV_ADD_WEIGHTED | ON | fusion of two images |
📃 WITH_FCV_EXTRACT_CHANNEL | ON | extract the data of the specified channel |
🗂️ BUILD_FCV_IMG_CALCULATION | ON | calculate from image |
📃 WITH_FCV_MATRIX_MUL | ON | matrix multiplication |
📃 WITH_FCV_NORM | ON | calculate image norm |
📃 WITH_FCV_MEAN | ON | calculate image mean |
📃 WITH_FCV_MIN_AREA_RECT | ON | generate the min bounding rectangle |
📃 WITH_FCV_MIN_MAX_LOC | ON | get the minimum and maximum coordinate points |
📃 WITH_FCV_CONNECTED_COMPONENTS | ON | calculate the connected domain |
📃 WITH_FCV_FIND_HOMOGRAPHY | ON | calculate the optimal single-map transformation |
🗂️ BUILD_FCV_FUSION_API | ON | fusion of image processing |
📃 WITH_FCV_BGR_TO_RGBA_WITH_MASK | ON | convert bgr to bgra with mask |
📃 WITH_FCV_NORMALIZE_TO_SUBMEAN_TO_REORDER | ON | image normalize & submean & reorder |
📃 WITH_FCV_SPLIT_TO_MEMCPY | ON | split image and copy to the specific memory |
📃 WITH_FCV_Y420SP_TO_RESIZE_TO_BGR | ON | convert yuv to bgr with resize |
📃 WITH_FCV_BGRA_TO_RESIZE_TO_BGR | ON | convert bgra to bgr with resize |
🗂️ BUILD_FCV_MATH_OPERATOR | ON | base math operators |
📃 WITH_FCV_VECTOR_OPERATOR | ON | vector operators |
🗂️ BUILD_FCV_MEDIA_IO | ON | media io |
📃 WITH_FCV_IMGCODECS | ON | image codesc |
📃 WITH_LIB_JPEG_TURBO | ON | support jpg codesc |
📃 WITH_LIB_PNG | ON | support png codesc |
🗂️ BUILD_FCV_IMG_DRAW | ON | drawing on image |
📃 WITH_FCV_FILL_POLY | ON | polygon drawing and filling |
📃 WITH_FCV_POLY_LINES | ON | polygon drawing |
📃 WITH_FCV_CIRCLE | ON | circle drawing |
📃 WITH_FCV_LINE | ON | line drawing |
compile options | default | description |
---|---|---|
🚀 ENABLE_SSE | ON | enable SSE instruction set optimization on x86 architecture |
🚀 ENABLE_AVX | ON | enable AVX instruction set optimization on x86 architecture |
🚀 ANDROID_ARM_NEON | ON | enable neon instruction set optimization on arm architecture |
🚀 ENABLE_NEON | ON | enable neon instruction set optimization on arm architecture |
🚀 ENABLE_SVE2 | OFF | enable SVE2 instruction set optimization on arm architecture |
compile options | default | description |
---|---|---|
BUILD_C | OFF | enable C interface |
compile options | default | description |
---|---|---|
🔖 BUILD_TEST | ON | enable unit test |
🔖 BUILD_BENCHMARK | ON | enable performance test |
🔖 BUILD_SAMPLES | ON | enable samples |