Cross-platform, customizable ML solutions for live and streaming media.
Go to file
2022-05-17 17:27:54 +05:30
.github Project import generated by Copybara. 2022-03-22 17:48:17 -07:00
docs Project import generated by Copybara. 2022-05-05 19:57:20 +00:00
mediapipe updated README, changed calculator name 2022-05-17 17:27:54 +05:30
third_party Project import generated by Copybara. 2022-05-05 19:57:20 +00:00
.bazelrc Project import generated by Copybara. 2022-05-05 19:57:20 +00:00
.bazelversion Project import generated by Copybara. 2022-03-21 12:12:39 -07:00
.dockerignore Project import generated by Copybara. 2019-06-16 16:06:57 -07:00
.gitignore Project import generated by Copybara. 2020-12-09 22:43:33 -05:00
build_android_examples.sh Project import generated by Copybara. 2022-03-21 12:12:39 -07:00
build_desktop_examples.sh Project import generated by Copybara. 2022-03-21 12:12:39 -07:00
build_ios_examples.sh Project import generated by Copybara. 2022-03-21 12:12:39 -07:00
BUILD.bazel Project import generated by Copybara. 2020-08-05 20:27:31 -04:00
CONTRIBUTING.md Project import generated by Copybara. 2021-12-13 15:56:02 -08:00
Dockerfile Project import generated by Copybara. 2022-03-21 12:12:39 -07:00
LICENSE Project import generated by Copybara. 2019-06-16 16:06:57 -07:00
MANIFEST.in Project import generated by Copybara. 2021-06-24 17:55:26 -04:00
README.md updated README, changed calculator name 2022-05-17 17:27:54 +05:30
requirements.txt Project import generated by Copybara. 2021-12-13 15:56:02 -08:00
setup_android_sdk_and_ndk.sh Project import generated by Copybara. 2022-05-05 19:57:20 +00:00
setup_opencv.sh Project import generated by Copybara. 2021-09-01 18:15:31 -07:00
setup.py Project import generated by Copybara. 2022-05-05 19:57:20 +00:00
WORKSPACE Project import generated by Copybara. 2022-05-05 19:57:20 +00:00

layout title nav_order
default Home 1

MediaPipe


Live ML anywhere

MediaPipe offers cross-platform, customizable ML solutions for live and streaming media.

accelerated.png cross_platform.png
End-to-End acceleration: Built-in fast ML inference and processing accelerated even on common hardware Build once, deploy anywhere: Unified solution works across Android, iOS, desktop/cloud, web and IoT
ready_to_use.png open_source.png
Ready-to-use solutions: Cutting-edge ML solutions demonstrating full power of the framework Free and open source: Framework and solutions both under Apache 2.0, fully extensible and customizable

ML solutions in MediaPipe

Face Detection Face Mesh Iris Hands Pose Holistic
face_detection face_mesh iris hand pose hair_segmentation
Hair Segmentation Object Detection Box Tracking Instant Motion Tracking Objectron KNIFT
hair_segmentation object_detection box_tracking instant_motion_tracking objectron knift
Android iOS C++ Python JS Coral
Face Detection
Face Mesh
Iris
Hands
Pose
Holistic
Selfie Segmentation
Hair Segmentation
Object Detection
Box Tracking
Instant Motion Tracking
Objectron
KNIFT
AutoFlip
MediaSequence
YouTube 8M

See also MediaPipe Models and Model Cards for ML models released in MediaPipe.

Getting started

To start using MediaPipe solutions with only a few lines code, see example code and demos in MediaPipe in Python and MediaPipe in JavaScript.

To use MediaPipe in C++, Android and iOS, which allow further customization of the solutions as well as building your own, learn how to install MediaPipe and start building example applications in C++, Android and iOS.

The source code is hosted in the MediaPipe Github repository, and you can run code search using Google Open Source Code Search.

Publications

Videos

Events

Community

Alpha disclaimer

MediaPipe is currently in alpha at v0.7. We may be still making breaking API changes and expect to get to stable APIs by v1.0.

Contributing

We welcome contributions. Please follow these guidelines.

We use GitHub issues for tracking requests and bugs. Please post questions to the MediaPipe Stack Overflow with a mediapipe tag.

Steps to add an image_flip_calculator

Objective of this exercise is to create a custom calculator which can do some custom prcessing of the image('flipping') and then insert this calculator in the mediapipe graph pipe line. Description: Calculators are the blocks of code which reperesent a node in a subgraph in mediapipe computational graphs.

  1. Goto /mediapipe/calculators/image directory , which contains several image manipulation calculators, place your custom calculator code in a .cc file right here.
  2. Edit the BUILD file in the same image folder by creating a <cc_library> entry for your calculator (the name is 'custom_calculator' in this case, mentioning the name the calculator src file in last step in the section).
  3. Edit the /mediapipe/modules/pose_landmark/pose_landmark_cpu.pbtxt , creating a entry for your calculator by giving it a name ('CustomImageFlipCalculator' in this case, same as that of the class defined by the calculator in step 1. ) whose input willbe the inputstream of the graph and output will connect to the node previously connected to the original input stream)
  4. The node created in the step 3 will connect to you custom_calculator by making an entry in the BUILD file in /mediapipe/modules/pose_landmark/ ; edit the file by finding the entry which mentions name in the <register_as> section. is the name of the sub_graph described by the pose_landmark_cpu.pbtxt. In the 'deps' section, add the calculator entry as dependency to the subgraph(in this case "//mediapipe/calculators/image:custom_calculator").
  5. BUILD your custom calculator by running "bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/calculators/image:custom_calculator" , and also for the pose_detection by running "bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/examples/desktop/pose_tracking:pose_tracking_cpu", and then run the pose detection desktop example.