MediaPipe Framework in C++

Please follow instructions below to build C++ command-line example apps with MediaPipe Framework. To learn more about these example apps, start from Hello World! in C++.

Building C++ command-line example apps

You can compile and run desktop MediaPipe graphs using either CPU execution or GPU acceleration.

Option 1: Running on CPU

1. To build, for example, the C++ hand tracking example graph on CPU, run:

   bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 \
   mediapipe/examples/desktop/hand_tracking:hand_tracking_cpu
   

2. To run the application:

   GLOG_logtostderr=1 bazel-bin/mediapipe/examples/desktop/hand_tracking/hand_tracking_cpu \
     --calculator_graph_config_file=mediapipe/graphs/hand_tracking/hand_tracking_desktop_live.pbtxt
   

   This will open up your webcam as long as it is connected and on. Any errors is likely due to your webcam being not accessible.

Option 2: Running on GPU

For Linux

First follow OpenGL ES Setup on Linux Desktop.

1. To build, for example, the hand tracking GPU example, run:

   bazel build -c opt --copt -DMESA_EGL_NO_X11_HEADERS --copt -DEGL_NO_X11 \
     mediapipe/examples/desktop/hand_tracking:hand_tracking_gpu
   

For macOS

1. To build, for example, the hand tracking GPU example, run:

   bazel build -c opt \
     --copt=-DMEDIAPIPE_METAL_ENABLED=1 \
     --copt=-DMEDIAPIPE_DISABLE_GPU=0 \
     --copt=-DMEDIAPIPE_GPU_BUFFER_USE_CV_PIXEL_BUFFER=1 \
     --copt=-DOSX_ENABLE_3_2_CORE \
     mediapipe/examples/desktop/hand_tracking:hand_tracking_gpu
   

Running the application

1. To run the application:

   GLOG_logtostderr=1 bazel-bin/mediapipe/examples/desktop/hand_tracking/hand_tracking_gpu \
     --calculator_graph_config_file=mediapipe/graphs/hand_tracking/hand_tracking_desktop_live_gpu.pbtxt
   

   This will open up your webcam as long as it is connected and on. Any errors is likely due to your webcam being not accessible, or GPU drivers not setup properly.