Computer Vision & Media Build Guide

OpenCV Compilation & FFmpeg Integration

A complete guide to compiling OpenCV 4.x from source using CMake across Windows, macOS, and Linux, and seamlessly hooking it into FFmpeg to unlock powerful computer vision video filters.

Step 1: Source Code & CMake Toolchain

Download OpenCV and prepare the cross-platform CMake build system

OpenCV 4.x Source

The open-source computer vision library. We use the 4.x branch for modern compatibility and stability.

Website Download 4.x

Extraction Command:

unzip opencv-4.x.zip && cd opencv-4.x

CMake (Required)

Unlike FFmpeg's configure script, OpenCV heavily relies on CMake to generate platform-specific makefiles.

Website

Installation:

# Windows (MSYS2): pacman -S mingw-w64-x86_64-cmake
# Ubuntu/Linux: sudo apt install cmake build-essential
# macOS: brew install cmake

Step 2: Cross-Platform OpenCV Build

Crucial: You MUST enable OPENCV_GENERATE_PKGCONFIG for FFmpeg to find it later.

The PKG-CONFIG Trap

FFmpeg's configure script uses `pkg-config` to check if OpenCV exists. By default, OpenCV 4's CMake does NOT generate a .pc file. You must explicitly pass `-DOPENCV_GENERATE_PKGCONFIG=ON` during the cmake phase.

1. Windows Platform (Run inside MSYS2 MINGW64 shell)

cd opencv-4.x
mkdir build && cd build

# Windows MinGW requires specific generator
cmake -G "MinGW Makefiles" \
    -DOPENCV_GENERATE_PKGCONFIG=ON \
    -DCMAKE_INSTALL_PREFIX=/usr/local/opencv4 \
    -DBUILD_SHARED_LIBS=ON \
    -DBUILD_TESTS=OFF \
    -DBUILD_PERF_TESTS=OFF \
    ..

mingw32-make -j$(nproc)
mingw32-make install

2. Ubuntu / Debian Linux Platform

cd opencv-4.x
mkdir build && cd build

cmake -DOPENCV_GENERATE_PKGCONFIG=ON \
    -DCMAKE_INSTALL_PREFIX=/usr/local/opencv4 \
    -DBUILD_SHARED_LIBS=ON \
    -DBUILD_TESTS=OFF \
    ..

make -j$(nproc)
sudo make install

3. macOS Platform

cd opencv-4.x
mkdir build && cd build

cmake -DOPENCV_GENERATE_PKGCONFIG=ON \
    -DCMAKE_INSTALL_PREFIX=/usr/local/opencv4 \
    -DWITH_OPENJPEG=OFF \
    -DBUILD_TESTS=OFF \
    ..

make -j$(sysctl -n hw.logicalcpu)
make install

Step 3: Integrate into FFmpeg

Export the pkg-config path and recompile FFmpeg with --enable-libopencv

link_ffmpeg.sh

# 1. Expose OpenCV's .pc file to the system PKG_CONFIG_PATH
export PKG_CONFIG_PATH="/usr/local/opencv4/lib/pkgconfig:$PKG_CONFIG_PATH"

# 2. Verify pkg-config can find it (Should output OpenCV flags)
pkg-config --cflags --libs opencv4

# 3. Go to FFmpeg source and configure with OpenCV enabled
cd ../ffmpeg-7.0
./configure \
    --enable-libopencv \
    --enable-gpl \
    --enable-shared \
    --extra-cflags="-I/usr/local/opencv4/include/opencv4"
make -j$(nproc) && make install

Step 4: Verification & Usage Examples

Confirm the filters are loaded and run an edge detection test

1. Verify OpenCV Filters

Check if FFmpeg correctly registered the OpenCV video filters (edgedetect, facedetect, etc.).

Console Command:

./ffmpeg -filters | grep opencv

Expected Output Match:

... edgedetect V->V Detect and draw edge. (opencv)
... facedetect V->V Detect and draw faces. (opencv)

2. Run Edge Detection Test

Apply the Canny edge detector from OpenCV directly to a video stream.

CLI Execution:

./ffmpeg -i input.mp4 -vf "edgedetect=low=0.1:high=0.4" -c:a copy output_edged.mp4

Explore More#opencv-dev

Cross-Platform OpenCV Compilation and Deep Integration with FFmpeg

Article

A detailed guide on how to compile OpenCV 4.x cleanly using CMake across multiple platforms, resolve core pkg-config linking issues, and link it into FFmpeg to unlock powerful video vision processing filters.

OpenCV Image Duplicate Detection and Similarity Analysis

Article

A detailed demonstration of how to use OpenCV for image duplicate detection and similarity analysis, including feature extraction, matching algorithms, and result visualization.

Python Calling OpenCV Image Sharpening

Article

A detailed demonstration of how to use Python to call OpenCV for image sharpening, enhancing image clarity and detail.

C++ Calling OpenCV Image Sharpening

Article

A detailed demonstration of how to use C++ to call OpenCV for image sharpening, enhancing image clarity and detail.

C++ OpenCV + FFmpeg 7: Making Videos Clearer from Blurry

Article

A detailed demonstration of how to use C++ and FFmpeg 7 in conjunction with OpenCV for video processing, achieving clarity from blurriness.