### Install Python Dependencies

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Setup Python 3 and the required six library.

```default
$ brew install python
$ sudo ln -s -f /usr/local/bin/python3.7 /usr/local/bin/python
$ python --version
Python 3.7.4
$ pip3 install --user six
```

--------------------------------

### Build GPU Examples with Custom Copts

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Builds MediaPipe GPU examples on Linux desktop by overriding default GPU disable flags. Use when Mesa GPU libraries are installed.

```bash
# To compile with GPU support, replace
--define MEDIAPIPE_DISABLE_GPU=1
# with
--copt -DMESA_EGL_NO_X11_HEADERS --copt -DEGL_NO_X11
# when building GPU examples.
```

--------------------------------

### Run YouTube-8M Inference Web Server

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/youtube_8m.md

Starts a Python web server for the YouTube-8M inference viewer. Requires 'absl-py' to be installed.

```bash
python mediapipe/examples/desktop/youtube8m/viewer/server.py --root `pwd`
```

--------------------------------

### Configure Custom OpenCV Path

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Example configuration for WORKSPACE and BUILD files when using a custom OpenCV installation path.

```bash
new_local_repository(
    name = "linux_opencv",
    build_file = "@//third_party:opencv_linux.BUILD",
    path = "/usr/local",
)

cc_library(
    name = "opencv",
    srcs = glob(
        [
            "lib/libopencv_core.so",
            "lib/libopencv_highgui.so",
            "lib/libopencv_imgcodecs.so",
            "lib/libopencv_imgproc.so",
            "lib/libopencv_video.so",
            "lib/libopencv_videoio.so",
        ],
    ),
    hdrs = glob(["include/opencv4/**/*.h*"]),
    includes = ["include/opencv4/"],
    linkstatic = 1,
    visibility = ["//visibility:public"],
)
```

--------------------------------

### Build and Install MediaPipe Package

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python.md

Build and install the MediaPipe package from source.

```bash
(mp_env)mediapipe$ python3 setup.py install --link-opencv
```

```bash
(mp_env)mediapipe$ python3 setup.py bdist_wheel
```

--------------------------------

### Run Hello World C++ Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Execute the MediaPipe Hello World example using Bazel with CPU-only configuration.

```bash
$ export GLOG_logtostderr=1
# Need bazel flag 'MEDIAPIPE_DISABLE_GPU=1' if you are running on Linux desktop with CPU only
$ bazel run --define MEDIAPIPE_DISABLE_GPU=1 \
    mediapipe/examples/desktop/hello_world:hello_world

# Should print:
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
```

--------------------------------

### Start Graph and Camera

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/hello_world_ios.md

Initialize the graph and start the camera source after permission is granted.

```objective-c
[_cameraSource requestCameraAccessWithCompletionHandler:^void(BOOL granted) {
  if (granted) {
    // Start running self.mediapipeGraph.
    NSError* error;
    if (![self.mediapipeGraph startWithError:&error]) {
      NSLog(@"Failed to start graph: %@", error);
    }
    else if (![self.mediapipeGraph waitUntilIdleWithError:&error]) {
      NSLog(@"Failed to complete graph initial run: %@", error);
    }

    dispatch_async(_videoQueue, ^{
      [_cameraSource start];
    });
  }
}];
```

--------------------------------

### Run Hello World C++ Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Executes the MediaPipe Hello World example with GPU support disabled.

```bash
username@DESKTOP-TMVLBJ1:~/mediapipe$ export GLOG_logtostderr=1

# Need bazel flag 'MEDIAPIPE_DISABLE_GPU=1' as desktop GPU is currently not supported
username@DESKTOP-TMVLBJ1:~/mediapipe$ bazel run --define MEDIAPIPE_DISABLE_GPU=1 \
    mediapipe/examples/desktop/hello_world:hello_world

# Should print:
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
```

--------------------------------

### Build and Run Hello World Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Build the C++ Hello World example using Bazel and execute the resulting binary.

```default
C:\Users\Username\mediapipe_repo>bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 --action_env PYTHON_BIN_PATH="C://python_36//python.exe" mediapipe/examples/desktop/hello_world

C:\Users\Username\mediapipe_repo>set GLOG_logtostderr=1

C:\Users\Username\mediapipe_repo>bazel-bin\mediapipe\examples\desktop\hello_world\hello_world.exe

# should print:
# I20200514 20:43:12.277598  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.278597  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.279618  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.279618  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.279618  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.279618  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.279618  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.279618  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.279618  1200 hello_world.cc:56] Hello World!
# I20200514 20:43:12.280613  1200 hello_world.cc:56] Hello World!
```

--------------------------------

### Run Hello World C++ Example in Docker

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Executes the desktop Hello World example within the running container.

```bash
$ docker run -it --name mediapipe mediapipe:latest

root@bca08b91ff63:/mediapipe# GLOG_logtostderr=1 bazel run --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/examples/desktop/hello_world
```

--------------------------------

### Clone and Run MediaPipe Hello World Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/hello_world_cpp.md

Clone the MediaPipe repository and run the 'hello_world' example using Bazel. Ensure GPU is disabled for desktop execution.

```bash
$ git clone https://github.com/google/mediapipe.git
$ cd mediapipe

$ export GLOG_logtostderr=1
# Need bazel flag 'MEDIAPIPE_DISABLE_GPU=1' as desktop GPU is not supported currently.
$ bazel run --define MEDIAPIPE_DISABLE_GPU=1 \
    mediapipe/examples/desktop/hello_world:hello_world

# It should print 10 rows of Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
```

--------------------------------

### Run C++ Hello World Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Execute the MediaPipe Hello World example using Bazel with GPU support disabled.

```bash
$ export GLOG_logtostderr=1
# Need bazel flag 'MEDIAPIPE_DISABLE_GPU=1' as desktop GPU is currently not supported
$ bazel run --define MEDIAPIPE_DISABLE_GPU=1 \
    mediapipe/examples/desktop/hello_world:hello_world

# Should print:
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
```

--------------------------------

### Install OpenCV Core Development Libraries

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Install essential OpenCV development libraries using apt-get. This command installs core, highgui, calib3d, features2d, imgproc, and video modules.

```bash
sudo apt-get install -y \
    libopencv-core-dev \
    libopencv-highgui-dev \
    libopencv-calib3d-dev \
    libopencv-features2d-dev \
    libopencv-imgproc-dev \
    libopencv-video-dev
```

--------------------------------

### Install Bazelisk via Homebrew

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/ios.md

Uses Homebrew to install the Bazelisk build tool.

```bash
brew install bazelisk
```

--------------------------------

### Install MediaPipe Package

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python.md

Install the MediaPipe package via pip and launch the Python interpreter.

```bash
(mp_env)$ pip install mediapipe
(mp_env)$ python3
```

--------------------------------

### Install Required Packages

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Installs essential build tools, git, python, zip, adb, and Java JDK.

```bash
username@DESKTOP-TMVLBJ1:~$ sudo apt-get update && sudo apt-get install -y build-essential git python zip adb openjdk-8-jdk
```

--------------------------------

### Install OpenCV Libraries

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Installs pre-compiled OpenCV development libraries using the apt package manager.

```bash
username@DESKTOP-TMVLBJ1:~/mediapipe$ sudo apt-get install libopencv-core-dev libopencv-highgui-dev \
                       libopencv-calib3d-dev libopencv-features2d-dev \
                       libopencv-imgproc-dev libopencv-video-dev
```

--------------------------------

### Build Android Example in Docker

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Sets up the Android SDK/NDK and builds the object detection Android application.

```bash
$ docker run -it --name mediapipe mediapipe:latest

root@bca08b91ff63:/mediapipe# bash ./setup_android_sdk_and_ndk.sh

root@bca08b91ff63:/mediapipe# bazel build -c opt --config=android_arm64 mediapipe/examples/android/src/java/com/google/mediapipe/apps/objectdetectiongpu:objectdetectiongpu
```

--------------------------------

### Build and Install Android Application

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/hello_world_android.md

Commands to build the APK using Bazel and install it on a connected device via ADB.

```default
bazel build -c opt --config=android_arm64 $APPLICATION_PATH:helloworld
```

```default
adb install bazel-bin/$APPLICATION_PATH/helloworld.apk
```

--------------------------------

### Setup Python Virtual Environment

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python.md

Create and activate a virtual environment for MediaPipe development.

```bash
$ python3 -m venv mp_env && source mp_env/bin/activate
```

--------------------------------

### Install Dependencies on Debian or Ubuntu

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python.md

Install required system dependencies for building MediaPipe on Debian-based systems.

```bash
$ sudo apt install python3-dev
$ sudo apt install python3-venv
$ sudo apt install -y protobuf-compiler

# If you need to build opencv from source.
$ sudo apt install cmake
```

--------------------------------

### Start Camera with Metadata Selection

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/hello_world_android.md

Use the retrieved metadata to determine the camera facing direction before starting the camera.

```java
CameraHelper.CameraFacing cameraFacing =
    applicationInfo.metaData.getBoolean("cameraFacingFront", false)
        ? CameraHelper.CameraFacing.FRONT
        : CameraHelper.CameraFacing.BACK;
cameraHelper.startCamera(this, cameraFacing, /*unusedSurfaceTexture=*/ null);
```

--------------------------------

### Install MSYS2 Packages

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Install essential build tools using the pacman package manager.

```default
C:\> pacman -S git patch unzip
```

--------------------------------

### Start Graph and Add Packets

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python_framework.md

Start the graph execution and inject data packets into input streams.

```python
graph.start_run()

graph.add_packet_to_input_stream(
    'in_stream', mp.packet_creator.create_string('abc').at(0))

rgb_img = cv2.cvtColor(cv2.imread('/path/to/your/image.png'), cv2.COLOR_BGR2RGB)
graph.add_packet_to_input_stream(
    'in_stream',
    mp.packet_creator.create_image_frame(image_format=mp.ImageFormat.SRGB,
                                         data=rgb_img).at(1))
```

--------------------------------

### Build Objectron Desktop Application

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/objectron.md

Command to compile the Objectron CPU example using Bazel.

```bash
bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/examples/desktop/object_detection_3d:objectron_cpu
```

--------------------------------

### Install Xcode Command Line Tools

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/ios.md

Installs the necessary command line tools for Xcode development.

```bash
xcode-select --install
```

--------------------------------

### Install msvc-runtime package

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/troubleshooting.md

Use this command to install the msvc-runtime package if you encounter DLL load failures on Windows.

```bash
$ python -m pip install msvc-runtime
```

--------------------------------

### Verify CUPTI Installation

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/gpu_support.md

List the contents of the CUPTI directory to confirm its installation. This is a component required for CUDA.

```bash
$ ls /usr/local/cuda/extras/CUPTI
/lib64
libcupti.so       libcupti.so.10.1.208  libnvperf_host.so        libnvperf_target.so
libcupti.so.10.1  libcupti_static.a     libnvperf_host_static.a
```

--------------------------------

### Hello World Output

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Expected output when running the MediaPipe C++ Hello World example.

```text
# Should print:
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
# Hello World!
```

--------------------------------

### Build TensorFlow Object Detection Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/object_detection.md

Compiles the TensorFlow CPU inference example for desktop. Note that the initial build may take significant time as it builds TensorFlow targets from scratch.

```bash
bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 --define no_aws_support=true --linkopt=-s \
mediapipe/examples/desktop/object_detection:object_detection_tensorflow
```

--------------------------------

### Compile Face Detection Coral Example with Bazel

Source: https://github.com/google-ai-edge/mediapipe/blob/master/mediapipe/examples/coral/README.md

Compiles the MediaPipe face detection example for Coral USB devices using Bazel. Requires libusb to be installed. Sets optimization mode, disables GPU, and specifies Edge TPU type.

```bash
bazel build \
  --compilation_mode=opt \
  --define darwinn_portable=1 \
  --define MEDIAPIPE_DISABLE_GPU=1 \
  --define MEDIAPIPE_EDGE_TPU=usb \
  --linkopt=-l:libusb-1.0.so \
  mediapipe/examples/coral:face_detection_tpu build
```

--------------------------------

### Run YouTube-8M Web Interface Server

Source: https://github.com/google-ai-edge/mediapipe/blob/master/mediapipe/examples/desktop/youtube8m/README.md

Starts the Python web server for the YouTube-8M inference viewer. Requires 'absl-py' to be installed. Navigate to localhost:8008 in a browser to access the interface.

```bash
python mediapipe/examples/desktop/youtube8m/viewer/server.py --root `pwd`
```

--------------------------------

### Install MediaPipe Holistic with NPM

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/javascript.md

Use npm to install the MediaPipe Holistic package locally. This command installs the specified package and its dependencies.

```bash
npm install @mediapipe/holistic.
```

--------------------------------

### Run MediaPipe Hello World (CPU)

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Executes the C++ Hello World example on Linux desktop using CPU only. Sets GLOG to log to stderr.

```bash
export GLOG_logtostderr=1

# if you are running on Linux desktop with CPU only
$ bazel run --define MEDIAPIPE_DISABLE_GPU=1 \
    mediapipe/examples/desktop/hello_world:hello_world
```

--------------------------------

### Face Effect Mobile iOS Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/face_mesh.md

Face effect example for iOS. This example works for a single face and requires a specific build target.

```bash
mediapipe/examples/ios/faceeffect
```

--------------------------------

### Face Effect Mobile Android Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/face_mesh.md

Face effect example for Android. This example works for a single face and requires a specific build target.

```bash
mediapipe/examples/android/src/java/com/google/mediapipe/apps/faceeffect
```

--------------------------------

### Build MediaPipe Hands for CPU

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/cpp.md

Compiles the hand tracking example for CPU execution using Bazel.

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

--------------------------------

### Face Mesh Desktop CPU Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/face_mesh.md

Run Face Mesh on desktop using CPU. The graph file and target build are specified.

```bash
mediapipe/examples/desktop/face_mesh:face_mesh_cpu
```

--------------------------------

### Install Dependencies on macOS

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python.md

Install required system dependencies for building MediaPipe on macOS.

```bash
$ brew install protobuf

# If you need to build opencv from source.
$ brew install cmake
```

--------------------------------

### Run MediaPipe Hello World (GPU)

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Executes the C++ Hello World example on Linux desktop with GPU support enabled via Mesa drivers. Sets GLOG to log to stderr.

```bash
export GLOG_logtostderr=1

# If you are running on Linux desktop with GPU support enabled (via mesa drivers)
$ bazel run --copt -DMESA_EGL_NO_X11_HEADERS --copt -DEGL_NO_X11 \
    mediapipe/examples/desktop/hello_world:hello_world
```

--------------------------------

### Install Python Six Library

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/ios.md

Installs the six library required for TensorFlow compatibility.

```bash
pip3 install --user six
```

--------------------------------

### Face Mesh Desktop GPU Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/face_mesh.md

Run Face Mesh on desktop using GPU. The graph file and target build are specified.

```bash
mediapipe/examples/desktop/face_mesh:face_mesh_gpu
```

--------------------------------

### Build MediaPipe Hands for GPU

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/cpp.md

Compiles the hand tracking example for GPU execution on Linux using Bazel.

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

--------------------------------

### Install OpenCV via MacPorts

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Alternative installation method for OpenCV using the MacPorts package manager.

```bash
$ port install opencv
```

--------------------------------

### Initialize and configure CameraXPreviewHelper

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/hello_world_android.md

Initialize CameraXPreviewHelper and set a listener to handle camera start events. This listener receives the SurfaceTexture for displaying frames and makes the preview view visible.

```java
public void startCamera() {
  cameraHelper = new CameraXPreviewHelper();
  cameraHelper.setOnCameraStartedListener(
    surfaceTexture -> {
      previewFrameTexture = surfaceTexture;
      // Make the display view visible to start showing the preview.
      previewDisplayView.setVisibility(View.VISIBLE);
    });
}
```

--------------------------------

### Install OpenCV via Package Manager

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Use yum to install the pre-compiled OpenCV development files.

```bash
$ sudo yum install opencv-devel
```

--------------------------------

### Import MediaPipe Solution

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python.md

Import the MediaPipe package and initialize a specific solution.

```python
import mediapipe as mp
mp_face_mesh = mp.solutions.face_mesh
```

--------------------------------

### Install Python Requirements

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python.md

Install necessary Python packages from the requirements file within the virtual environment.

```bash
(mp_env)mediapipe$ pip3 install -r requirements.txt
```

--------------------------------

### Build Template Matching Index File

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/knift.md

Builds the template matching application for desktop and generates an index file for template images. Ensure all template images are in a single directory.

```bash
bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 \
  mediapipe/examples/desktop/template_matching:template_matching_tflite
```

```bash
bazel-bin/mediapipe/examples/desktop/template_matching/template_matching_tflite \
  --calculator_graph_config_file=mediapipe/graphs/template_matching/index_building.pbtxt \
  --input_side_packets="file_directory=<template image directory>,file_suffix=png,output_index_filename=<output index filename>"
```

--------------------------------

### Build the AutoFlip Pipeline

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/autoflip.md

Use Bazel to build the AutoFlip desktop example. Requires OpenCV 3.

```bash
bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/examples/desktop/autoflip:run_autoflip
```

--------------------------------

### Install OpenCV and FFmpeg via Homebrew

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/install.md

Use Homebrew to install OpenCV and FFmpeg, then remove the conflicting glog dependency.

```bash
$ brew install opencv@3

# There is a known issue caused by the glog dependency. Uninstall glog.
$ brew uninstall --ignore-dependencies glog
```

--------------------------------

### Verify NVCC Installation

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/gpu_support.md

Check the version of the NVIDIA CUDA Compiler (NVCC). This confirms that the CUDA compiler is installed and accessible.

```bash
$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Sun_Jul_28_19:07:16_PDT_2019
Cuda compilation tools, release 10.1, V10.1.243
```

--------------------------------

### Run the AutoFlip Pipeline

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/autoflip.md

Execute the binary with input/output paths and a target aspect ratio.

```bash
GLOG_logtostderr=1 bazel-bin/mediapipe/examples/desktop/autoflip/run_autoflip \
  --calculator_graph_config_file=mediapipe/examples/desktop/autoflip/autoflip_graph.pbtxt \
  --input_side_packets=input_video_path=/absolute/path/to/the/local/video/file,output_video_path=/absolute/path/to/save/the/output/video/file,aspect_ratio=1:1
```

--------------------------------

### Build TFLite Object Detection Example

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/object_detection.md

Compiles the TFLite object detection example for desktop with GPU support disabled.

```bash
bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/examples/desktop/object_detection:object_detection_tflite
```

--------------------------------

### Initialize MediaPipe Face Detection in Python

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/solutions/face_detection.md

Setup the face detection module and drawing utilities for processing images.

```python
import cv2
import mediapipe as mp
mp_face_detection = mp.solutions.face_detection
mp_drawing = mp.solutions.drawing_utils
```

--------------------------------

### Verify cuDNN Installation

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/gpu_support.md

List cuDNN library files to confirm its installation. cuDNN is a GPU-accelerated library for deep neural networks.

```bash
$ ls /usr/lib/x86_64-linux-gnu/ | grep libcudnn.so
libcudnn.so
libcudnn.so.7
libcudnn.so.7.6.4
```

--------------------------------

### Install Missing Python Packages

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/troubleshooting.md

If you encounter 'ImportError: No module named ...', install the missing package using pip or pip3.

```bash
pip install numpy
```

--------------------------------

### Initialize CalculatorGraph with Protobuf

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/getting_started/python_framework.md

Initialize a graph using a text-based CalculatorGraphConfig and register a callback to observe output streams.

```python
import mediapipe as mp

config_text = """
  input_stream: 'in_stream'
  output_stream: 'out_stream'
  node {
    calculator: 'PassThroughCalculator'
    input_stream: 'in_stream'
    output_stream: 'out_stream'
  }
"""
graph = mp.CalculatorGraph(graph_config=config_text)
output_packets = []
graph.observe_output_stream(
    'out_stream',
    lambda stream_name, packet:
        output_packets.append(mp.packet_getter.get_str(packet)))
```

--------------------------------

### Calculator Graph Configuration: Multiple Output Identifiers

Source: https://github.com/google-ai-edge/mediapipe/blob/master/docs/framework_concepts/calculators.md

Illustrates identifying output streams by tag name, index number, and a combination of both. This example shows how to connect video and multiple audio streams.

```proto
# Graph describing calculator SomeAudioVideoCalculator
node {
  calculator: "SomeAudioVideoCalculator"
  input_stream: "combined_input"
  output_stream: "VIDEO:video_stream"
  output_stream: "AUDIO:0:audio_left"
  output_stream: "AUDIO:1:audio_right"
}

node {
  calculator: "SomeAudioCalculator"
  input_stream: "audio_left"
  input_stream: "audio_right"
  output_stream: "audio_energy"
}
```