Project import generated by Copybara.

GitOrigin-RevId: f72a0f86c2c2acdb1920973c718a9e26ed3ec4b6
This commit is contained in:
MediaPipe Team 2020-06-05 19:49:27 -04:00 committed by chuoling
parent 5d028d923b
commit cd2b69d58c
315 changed files with 8162 additions and 7502 deletions

8
.gitignore vendored
View File

@ -1,8 +1,4 @@
mediapipe/provisioning_profile.mobileprovision
bazel-bin
bazel-genfiles
bazel-mediapipe-ioss
bazel-out
bazel-testlogs
bazel-*
mediapipe/MediaPipe.xcodeproj
mediapipe/MediaPipe.tulsiproj/*.tulsiconf-user
mediapipe/provisioning_profile.mobileprovision

167
README.md
View File

@ -1,84 +1,141 @@
![MediaPipe](mediapipe/docs/images/mediapipe_small.png?raw=true "MediaPipe logo")
=======================================================================
---
layout: default
title: Home
nav_order: 1
---
[MediaPipe](http://mediapipe.dev) is the simplest way for researchers and developers to build world-class ML solutions and applications for mobile, edge, cloud and the web.
![MediaPipe](docs/images/mediapipe_small.png)
## ML Solutions in MediaPipe
--------------------------------------------------------------------------------
* [Face Detection](mediapipe/docs/face_detection_mobile_gpu.md) [(web demo)](https://viz.mediapipe.dev/runner/demos/face_detection/face_detection.html)
* [Face Mesh](mediapipe/docs/face_mesh_mobile_gpu.md)
* [Hand Detection](mediapipe/docs/hand_detection_mobile_gpu.md)
* [Hand Tracking](mediapipe/docs/hand_tracking_mobile_gpu.md) [(web demo)](https://viz.mediapipe.dev/runner/demos/hand_tracking/hand_tracking.html)
* [Multi-hand Tracking](mediapipe/docs/multi_hand_tracking_mobile_gpu.md)
* [Hair Segmentation](mediapipe/docs/hair_segmentation_mobile_gpu.md) [(web demo)](https://viz.mediapipe.dev/runner/demos/hair_segmentation/hair_segmentation.html)
* [Object Detection](mediapipe/docs/object_detection_mobile_gpu.md)
* [Object Detection and Tracking](mediapipe/docs/object_tracking_mobile_gpu.md)
* [Objectron: 3D Object Detection and Tracking](mediapipe/docs/objectron_mobile_gpu.md)
* [AutoFlip: Intelligent Video Reframing](mediapipe/docs/autoflip.md)
* [KNIFT: Template Matching with Neural Image Features](mediapipe/docs/template_matching_mobile_cpu.md)
## Cross-platform ML solutions made simple
![face_detection](mediapipe/docs/images/mobile/face_detection_android_gpu_small.gif)
![face_mesh](mediapipe/docs/images/mobile/face_mesh_android_gpu_small.gif)
![hand_tracking](mediapipe/docs/images/mobile/hand_tracking_android_gpu_small.gif)
![multi-hand_tracking](mediapipe/docs/images/mobile/multi_hand_tracking_3d_android_gpu_small.gif)
![hair_segmentation](mediapipe/docs/images/mobile/hair_segmentation_android_gpu_small.gif)
![object_detection](mediapipe/docs/images/mobile/object_detection_android_gpu_small.gif)
![object_tracking](mediapipe/docs/images/mobile/object_tracking_android_gpu_small.gif)
![objectron_shoes](mediapipe/docs/images/mobile/objectron_shoe_android_gpu_small.gif)
![objectron_chair](mediapipe/docs/images/mobile/objectron_chair_android_gpu_small.gif)
![template_matching](mediapipe/docs/images/mobile/template_matching_android_cpu_small.gif)
[MediaPipe](https://google.github.io/mediapipe/) is the simplest way for researchers
and developers to build world-class ML solutions and applications for mobile,
desktop/cloud, web and IoT devices.
## Installation
Follow these [instructions](mediapipe/docs/install.md).
![accelerated.png](docs/images/accelerated_small.png) | ![cross_platform.png](docs/images/cross_platform_small.png)
:------------------------------------------------------------------------------------------------------------: | :----------------------------------------------------:
***End-to-End acceleration***: *built-in fast ML inference and processing accelerated even on common hardware* | ***Build one, deploy anywhere***: *Unified solution works across Android, iOS, desktop/cloud, web and IoT*
![ready_to_use.png](docs/images/ready_to_use_small.png) | ![open_source.png](docs/images/open_source_small.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 | Hand | Hair Segmentation
:----------------------------------------------------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------: | :---------------:
[![face_detection](docs/images/mobile/face_detection_android_gpu_small.gif)](https://google.github.io/mediapipe/solutions/face_detection) | [![face_mesh](docs/images/mobile/face_mesh_android_gpu_small.gif)](https://google.github.io/mediapipe/solutions/face_mesh) | [![hand](docs/images/mobile/hand_tracking_android_gpu_small.gif)](https://google.github.io/mediapipe/solutions/hand) | [![hair_segmentation](docs/images/mobile/hair_segmentation_android_gpu_small.gif)](https://google.github.io/mediapipe/solutions/hair_segmentation)
Object Detection | Box Tracking | Objectron | KNIFT
:----------------------------------------------------------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------------------: | :---:
[![object_detection](docs/images/mobile/object_detection_android_gpu_small.gif)](https://google.github.io/mediapipe/solutions/object_detection) | [![box_tracking](docs/images/mobile/object_tracking_android_gpu_small.gif)](https://google.github.io/mediapipe/solutions/box_tracking) | [![objectron](docs/images/mobile/objectron_chair_android_gpu_small.gif)](https://google.github.io/mediapipe/solutions/objectron) | [![knift](docs/images/mobile/template_matching_android_cpu_small.gif)](https://google.github.io/mediapipe/solutions/knift)
<!-- []() in the first cell is needed to preserve table formatting in GitHub Pages. -->
<!-- Whenever this table is updated, paste a copy to solutions/solutions.md. -->
[]() | Android | iOS | Desktop | Web | Coral
:---------------------------------------------------------------------------- | :-----: | :-: | :-----: | :-: | :---:
[Face Detection](https://google.github.io/mediapipe/solutions/face_detection) | ✅ | ✅ | ✅ | ✅ | ✅
[Face Mesh](https://google.github.io/mediapipe/solutions/face_mesh) | ✅ | ✅ | ✅ | |
[Hand](https://google.github.io/mediapipe/solutions/hand) | ✅ | ✅ | ✅ | ✅ |
[Hair Segmentation](https://google.github.io/mediapipe/solutions/hair_segmentation) | ✅ | | ✅ | ✅ |
[Object Detection](https://google.github.io/mediapipe/solutions/object_detection) | ✅ | ✅ | ✅ | | ✅
[Box Tracking](https://google.github.io/mediapipe/solutions/box_tracking) | ✅ | ✅ | ✅ | |
[Objectron](https://google.github.io/mediapipe/solutions/objectron) | ✅ | | | |
[KNIFT](https://google.github.io/mediapipe/solutions/knift) | ✅ | | | |
[AutoFlip](https://google.github.io/mediapipe/solutions/autoflip) | | | ✅ | |
[MediaSequence](https://google.github.io/mediapipe/solutions/media_sequence) | | | ✅ | |
[YouTube 8M](https://google.github.io/mediapipe/solutions/youtube_8m) | | | ✅ | |
## MediaPipe on the Web
MediaPipe on the Web is an effort to run the same ML solutions built for mobile
and desktop also in web browsers. The official API is under construction, but
the core technology has been proven effective. Please see
[MediaPipe on the Web](https://developers.googleblog.com/2020/01/mediapipe-on-web.html)
in Google Developers Blog for details.
You can use the following links to load a demo in the MediaPipe Visualizer, and
over there click the "Runner" icon in the top bar like shown below. The demos
use your webcam video as input, which is processed all locally in real-time and
never leaves your device.
![visualizer_runner](docs/images/visualizer_runner.png)
* [MediaPipe Face Detection](https://viz.mediapipe.dev/demo/face_detection)
* [MediaPipe Hand](https://viz.mediapipe.dev/demo/hand_tracking)
* [MediaPipe Hand (palm/hand detection only)](https://viz.mediapipe.dev/demo/hand_detection)
* [MediaPipe Hair Segmentation](https://viz.mediapipe.dev/demo/hair_segmentation)
## Getting started
See mobile, desktop, web and Google Coral [examples](mediapipe/docs/examples.md).
## Documentation
[MediaPipe Read-the-Docs](https://mediapipe.readthedocs.io/) or [docs.mediapipe.dev](https://docs.mediapipe.dev)
Learn how to [install](https://google.github.io/mediapipe/getting_started/install)
MediaPipe and
[build example applications](https://google.github.io/mediapipe/getting_started/building_examples),
and start exploring our ready-to-use
[solutions](https://google.github.io/mediapipe/solutions/solutions) that you can
further extend and customize.
Check out the [Examples page](https://mediapipe.readthedocs.io/en/latest/examples.html) for tutorials on how to use MediaPipe. [Concepts page](https://mediapipe.readthedocs.io/en/latest/concepts.html) for basic definitions
## Visualizing MediaPipe graphs
A web-based visualizer is hosted on [viz.mediapipe.dev](https://viz.mediapipe.dev/). Please also see instructions [here](mediapipe/docs/visualizer.md).
## Google Open Source Code search
Search MediaPipe Github repository using [Google Open Source code search](https://t.co/LSZnbMUUnT?amp=1)
## Videos
* [YouTube Channel](https://www.youtube.com/channel/UCObqmpuSMx-usADtL_qdMAw)
The source code is hosted in the
[MediaPipe Github repository](https://github.com/google/mediapipe), and you can
run code search using
[Google Open Source Code Search](https://cs.opensource.google/mediapipe/mediapipe).
## Publications
* [MediaPipe KNIFT: Template-based Feature Matching](https://mediapipe.page.link/knift-blog)
* [MediaPipe KNIFT: Template-based feature matching](https://developers.googleblog.com/2020/04/mediapipe-knift-template-based-feature-matching.html)
in Google Developers Blog
* [Alfred Camera: Smart camera features using MediaPipe](https://developers.googleblog.com/2020/03/alfred-camera-smart-camera-features-using-mediapipe.html)
* [MediaPipe Objectron: Real-time 3D Object Detection on Mobile Devices](https://mediapipe.page.link/objectron-aiblog)
* [AutoFlip: An Open Source Framework for Intelligent Video Reframing](https://mediapipe.page.link/autoflip)
* [Google Developer Blog: MediaPipe on the Web](https://mediapipe.page.link/webdevblog)
* [Google Developer Blog: Object Detection and Tracking using MediaPipe](https://mediapipe.page.link/objecttrackingblog)
in Google Developers Blog
* [Real-Time 3D Object Detection on Mobile Devices with MediaPipe](https://ai.googleblog.com/2020/03/real-time-3d-object-detection-on-mobile.html)
in Google AI Blog
* [AutoFlip: An Open Source Framework for Intelligent Video Reframing](https://ai.googleblog.com/2020/02/autoflip-open-source-framework-for.html)
in Google AI Blog
* [MediaPipe on the Web](https://developers.googleblog.com/2020/01/mediapipe-on-web.html)
in Google Developers Blog
* [Object Detection and Tracking using MediaPipe](https://developers.googleblog.com/2019/12/object-detection-and-tracking-using-mediapipe.html)
in Google Developers Blog
* [On-Device, Real-Time Hand Tracking with MediaPipe](https://ai.googleblog.com/2019/08/on-device-real-time-hand-tracking-with.html)
in Google AI Blog
* [MediaPipe: A Framework for Building Perception Pipelines](https://arxiv.org/abs/1906.08172)
## Videos
* [YouTube Channel](https://www.youtube.com/channel/UCObqmpuSMx-usADtL_qdMAw)
## Events
* [MediaPipe Seattle Meetup, Google Building Waterside, 13 Feb 2020](https://mediapipe.page.link/seattle2020)
* [AI Nextcon 2020, 12-16 Feb 2020, Seattle](http://aisea20.xnextcon.com/)
* [MediaPipe Madrid Meetup, 16 Dec 2019](https://www.meetup.com/Madrid-AI-Developers-Group/events/266329088/)
* [MediaPipe London Meetup, Google 123 Building, 12 Dec 2019](https://www.meetup.com/London-AI-Tech-Talk/events/266329038)
* [ML Conference, Berlin, 11 Dec 2019](https://mlconference.ai/machine-learning-advanced-development/mediapipe-building-real-time-cross-platform-mobile-web-edge-desktop-video-audio-ml-pipelines/)
* [MediaPipe Berlin Meetup, Google Berlin, 11 Dec 2019](https://www.meetup.com/Berlin-AI-Tech-Talk/events/266328794/)
* [The 3rd Workshop on YouTube-8M Large Scale Video Understanding Workshop](https://research.google.com/youtube8m/workshop2019/index.html) Seoul, Korea ICCV 2019
* [AI DevWorld 2019](https://aidevworld.com) on Oct 10 in San Jose, California
* [Google Industry Workshop at ICIP 2019](http://2019.ieeeicip.org/?action=page4&id=14#Google) [Presentation](https://docs.google.com/presentation/d/e/2PACX-1vRIBBbO_LO9v2YmvbHHEt1cwyqH6EjDxiILjuT0foXy1E7g6uyh4CesB2DkkEwlRDO9_lWfuKMZx98T/pub?start=false&loop=false&delayms=3000&slide=id.g556cc1a659_0_5) on Sept 24 in Taipei, Taiwan
* [Open sourced at CVPR 2019](https://sites.google.com/corp/view/perception-cv4arvr/mediapipe) on June 17~20 in Long Beach, CA
* [The 3rd Workshop on YouTube-8M Large Scale Video Understanding Workshop,
Seoul, Korea ICCV
2019](https://research.google.com/youtube8m/workshop2019/index.html)
* [AI DevWorld 2019, 10 Oct 2019, San Jose, CA](https://aidevworld.com)
* [Google Industry Workshop at ICIP 2019, 24 Sept 2019, Taipei, Taiwan](http://2019.ieeeicip.org/?action=page4&id=14#Google)
([presentation](https://docs.google.com/presentation/d/e/2PACX-1vRIBBbO_LO9v2YmvbHHEt1cwyqH6EjDxiILjuT0foXy1E7g6uyh4CesB2DkkEwlRDO9_lWfuKMZx98T/pub?start=false&loop=false&delayms=3000&slide=id.g556cc1a659_0_5))
* [Open sourced at CVPR 2019, 17~20 June, Long Beach, CA](https://sites.google.com/corp/view/perception-cv4arvr/mediapipe)
## Community
* [Awesome MediaPipe: curation of code related to MediaPipe](https://mediapipe.org)
* [Slack community for MediaPipe users](https://mediapipe.slack.com)
* [Discuss](https://groups.google.com/forum/#!forum/mediapipe) - General community discussion around MediaPipe
* [Awesome MediaPipe](https://mediapipe.org) - A curated list of awesome
MediaPipe related frameworks, libraries and software
* [Slack community](https://mediapipe.slack.com) for MediaPipe users
* [Discuss](https://groups.google.com/forum/#!forum/mediapipe) - General
community discussion around MediaPipe
## Alpha Disclaimer
MediaPipe is currently in alpha for v0.7. We are still making breaking API changes and expect to get to stable API by v1.0.
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](./CONTRIBUTING.md).
We use GitHub issues for tracking requests and bugs. Please post questions to the MediaPipe Stack Overflow with a 'mediapipe' tag.
We welcome contributions. Please follow these
[guidelines](https://github.com/google/mediapipe/blob/master/CONTRIBUTING.md).
We use GitHub issues for tracking requests and bugs. Please post questions to
the MediaPipe Stack Overflow with a `mediapipe` tag.

29
docs/_config.yml Normal file
View File

@ -0,0 +1,29 @@
# Configuration for GitHub Pages
remote_theme: pmarsceill/just-the-docs
# Set a path/url to a logo that will be displayed instead of the title
logo: "images/logo_horizontal_color.png"
# Enable or disable the site search
search_enabled: true
# Set the search token separator for hyphenated-word search:
search_tokenizer_separator: /[\s/]+/
# Enable or disable heading anchors
heading_anchors: true
# Aux links for the upper right navigation
aux_links:
"MediaPipe on GitHub":
- "//github.com/google/mediapipe"
# Footer content appears at the bottom of every page's main content
footer_content: "&copy; 2020 GOOGLE LLC | <a href=\"https://policies.google.com/privacy\">PRIVACY POLICY</a> | <a href=\"https://policies.google.com/terms\">TERMS OF SERVICE</a>"
# Color scheme currently only supports "dark" or nil (default)
color_scheme: nil
# Google Analytics Tracking (optional)
ga_tracking: UA-140696581-2

96
docs/examples.md Normal file
View File

@ -0,0 +1,96 @@
---
nav_exclude: true
---
# Examples
Below are code samples on how to run MediaPipe on both mobile and desktop. We
currently support MediaPipe APIs on mobile for Android only but will add support
for Objective-C shortly.
## Mobile
### [Hello World! on Android](./getting_started/hello_world_android.md)
This should be the first mobile Android example users go through in detail. It
teaches the following:
* Introduction of a simple MediaPipe graph running on mobile GPUs for
[Sobel edge detection](https://en.wikipedia.org/wiki/Sobel_operator).
* Building a simple baseline Android application that displays "Hello World!".
* Adding camera preview support into the baseline application using the
Android [CameraX] API.
* Incorporating the Sobel edge detection graph to process the live camera
preview and display the processed video in real-time.
[Sobel edge detection]:https://en.wikipedia.org/wiki/Sobel_operator
[CameraX]:https://developer.android.com/training/camerax
### [Hello World! on iOS](./getting_started/hello_world_ios.md)
This is the iOS version of Sobel edge detection example.
### [Face Detection](./solutions/face_detection.md)
### [Face Mesh](./solutions/face_mesh.md)
### [Hand](./solutions/hand.md)
### [Hair Segmentation](./solutions/hair_segmentation.md)
### [Object Detection](./solutions/object_detection.md)
### [Box Tracking](./solutions/box_tracking.md)
### [Objectron: 3D Object Detection](./solutions/objectron.md)
### [KNIFT: Template-based Feature Matching](./solutions/knift.md)
## Desktop
### [Hello World for C++](./getting_started/hello_world_desktop.md)
This shows how to run a simple graph using the MediaPipe C++ APIs.
### [Face Detection](./solutions/face_detection.md)
### [Face Mesh](./solutions/face_mesh.md)
### [Hand](./solutions/hand.md)
### [Hair Segmentation](./solutions/hair_segmentation.md)
### [Object Detection](./solutions/object_detection.md)
### [Box Tracking](./solutions/box_tracking.md)
### [AutoFlip - Semantic-aware Video Cropping](./solutions/autoflip.md)
### [Preparing Data Sets with MediaSequence](./solutions/media_sequence.md)
This shows how to use MediaPipe for media processing to prepare video data sets
for training a TensorFlow model.
### [Feature Extraction and Model Inference for YouTube-8M Challenge](./solutions/youtube_8m.md)
This shows how to use MediaPipe to prepare training data for the YouTube-8M
Challenge and do the model inference with the baseline model.
## Google Coral (ML acceleration with Google EdgeTPU)
### [Face Detection](./solutions/face_detection.md)
### [Object Detection](./solutions/object_detection.md)
## Web Browser
See more details [here](./getting_started/web.md) and
[Google Developer blog post](https://mediapipe.page.link/webdevblog).
### [Face Detection in Browser](https://viz.mediapipe.dev/demo/face_detection)
### [Hand Detection in Browser](https://viz.mediapipe.dev/demo/hand_detection)
### [Hand Tracking in Browser](https://viz.mediapipe.dev/demo/hand_tracking)
### [Hair Segmentation in Browser](https://viz.mediapipe.dev/demo/hair_segmentation)

View File

@ -0,0 +1,412 @@
---
layout: default
title: Calculators
parent: Framework Concepts
nav_order: 1
---
# Calculators
{: .no_toc }
1. TOC
{:toc}
---
Each calculator is a node of a graph. We describe how to create a new
calculator, how to initialize a calculator, how to perform its calculations,
input and output streams, timestamps, and options. Each node in the graph is
implemented as a `Calculator`. The bulk of graph execution happens inside its
calculators. A calculator may receive zero or more input streams and/or side
packets and produces zero or more output streams and/or side packets.
## CalculatorBase
A calculator is created by defining a new sub-class of the
[`CalculatorBase`](https://github.com/google/mediapipe/tree/master/mediapipe/framework/calculator_base.cc)
class, implementing a number of methods, and registering the new sub-class with
Mediapipe. At a minimum, a new calculator must implement the below four methods
* `GetContract()`
* Calculator authors can specify the expected types of inputs and outputs
of a calculator in GetContract(). When a graph is initialized, the
framework calls a static method to verify if the packet types of the
connected inputs and outputs match the information in this
specification.
* `Open()`
* After a graph starts, the framework calls `Open()`. The input side
packets are available to the calculator at this point. `Open()`
interprets the node configuration operations (see [Graphs](graphs.md))
and prepares the calculator's per-graph-run state. This function may
also write packets to calculator outputs. An error during `Open()` can
terminate the graph run.
* `Process()`
* For a calculator with inputs, the framework calls `Process()` repeatedly
whenever at least one input stream has a packet available. The framework
by default guarantees that all inputs have the same timestamp (see
[Synchronization](synchronization.md) for more information). Multiple
`Process()` calls can be invoked simultaneously when parallel execution
is enabled. If an error occurs during `Process()`, the framework calls
`Close()` and the graph run terminates.
* `Close()`
* After all calls to `Process()` finish or when all input streams close,
the framework calls `Close()`. This function is always called if
`Open()` was called and succeeded and even if the graph run terminated
because of an error. No inputs are available via any input streams
during `Close()`, but it still has access to input side packets and
therefore may write outputs. After `Close()` returns, the calculator
should be considered a dead node. The calculator object is destroyed as
soon as the graph finishes running.
The following are code snippets from
[CalculatorBase.h](https://github.com/google/mediapipe/tree/master/mediapipe/framework/calculator_base.h).
```c++
class CalculatorBase {
public:
...
// The subclasses of CalculatorBase must implement GetContract.
// ...
static ::MediaPipe::Status GetContract(CalculatorContract* cc);
// Open is called before any Process() calls, on a freshly constructed
// calculator. Subclasses may override this method to perform necessary
// setup, and possibly output Packets and/or set output streams' headers.
// ...
virtual ::MediaPipe::Status Open(CalculatorContext* cc) {
return ::MediaPipe::OkStatus();
}
// Processes the incoming inputs. May call the methods on cc to access
// inputs and produce outputs.
// ...
virtual ::MediaPipe::Status Process(CalculatorContext* cc) = 0;
// Is called if Open() was called and succeeded. Is called either
// immediately after processing is complete or after a graph run has ended
// (if an error occurred in the graph). ...
virtual ::MediaPipe::Status Close(CalculatorContext* cc) {
return ::MediaPipe::OkStatus();
}
...
};
```
## Life of a calculator
During initialization of a MediaPipe graph, the framework calls a
`GetContract()` static method to determine what kinds of packets are expected.
The framework constructs and destroys the entire calculator for each graph run
(e.g. once per video or once per image). Expensive or large objects that remain
constant across graph runs should be supplied as input side packets so the
calculations are not repeated on subsequent runs.
After initialization, for each run of the graph, the following sequence occurs:
* `Open()`
* `Process()` (repeatedly)
* `Close()`
The framework calls `Open()` to initialize the calculator. `Open()` should
interpret any options and set up the calculator's per-graph-run state. `Open()`
may obtain input side packets and write packets to calculator outputs. If
appropriate, it should call `SetOffset()` to reduce potential packet buffering
of input streams.
If an error occurs during `Open()` or `Process()` (as indicated by one of them
returning a non-`Ok` status), the graph run is terminated with no further calls
to the calculator's methods, and the calculator is destroyed.
For a calculator with inputs, the framework calls `Process()` whenever at least
one input has a packet available. The framework guarantees that inputs all have
the same timestamp, that timestamps increase with each call to `Process()` and
that all packets are delivered. As a consequence, some inputs may not have any
packets when `Process()` is called. An input whose packet is missing appears to
produce an empty packet (with no timestamp).
The framework calls `Close()` after all calls to `Process()`. All inputs will
have been exhausted, but `Close()` has access to input side packets and may
write outputs. After Close returns, the calculator is destroyed.
Calculators with no inputs are referred to as sources. A source calculator
continues to have `Process()` called as long as it returns an `Ok` status. A
source calculator indicates that it is exhausted by returning a stop status
(i.e. MediaPipe::tool::StatusStop).
## Identifying inputs and outputs
The public interface to a calculator consists of a set of input streams and
output streams. In a CalculatorGraphConfiguration, the outputs from some
calculators are connected to the inputs of other calculators using named
streams. Stream names are normally lowercase, while input and output tags are
normally UPPERCASE. In the example below, the output with tag name `VIDEO` is
connected to the input with tag name `VIDEO_IN` using the stream named
`video_stream`.
```proto
# Graph describing calculator SomeAudioVideoCalculator
node {
calculator: "SomeAudioVideoCalculator"
input_stream: "INPUT:combined_input"
output_stream: "VIDEO:video_stream"
}
node {
calculator: "SomeVideoCalculator"
input_stream: "VIDEO_IN:video_stream"
output_stream: "VIDEO_OUT:processed_video"
}
```
Input and output streams can be identified by index number, by tag name, or by a
combination of tag name and index number. You can see some examples of input and
output identifiers in the example below. `SomeAudioVideoCalculator` identifies
its video output by tag and its audio outputs by the combination of tag and
index. The input with tag `VIDEO` is connected to the stream named
`video_stream`. The outputs with tag `AUDIO` and indices `0` and `1` are
connected to the streams named `audio_left` and `audio_right`.
`SomeAudioCalculator` identifies its audio inputs by index only (no tag needed).
```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"
}
```
In the calculator implementation, inputs and outputs are also identified by tag
name and index number. In the function below input are output are identified:
* By index number: The combined input stream is identified simply by index
`0`.
* By tag name: The video output stream is identified by tag name "VIDEO".
* By tag name and index number: The output audio streams are identified by the
combination of the tag name `AUDIO` and the index numbers `0` and `1`.
```c++
// c++ Code snippet describing the SomeAudioVideoCalculator GetContract() method
class SomeAudioVideoCalculator : public CalculatorBase {
public:
static ::mediapipe::Status GetContract(CalculatorContract* cc) {
cc->Inputs().Index(0).SetAny();
// SetAny() is used to specify that whatever the type of the
// stream is, it's acceptable. This does not mean that any
// packet is acceptable. Packets in the stream still have a
// particular type. SetAny() has the same effect as explicitly
// setting the type to be the stream's type.
cc->Outputs().Tag("VIDEO").Set<ImageFrame>();
cc->Outputs().Get("AUDIO", 0).Set<Matrix>;
cc->Outputs().Get("AUDIO", 1).Set<Matrix>;
return ::mediapipe::OkStatus();
}
```
## Processing
`Process()` called on a non-source node must return `::mediapipe::OkStatus()` to
indicate that all went well, or any other status code to signal an error
If a non-source calculator returns `tool::StatusStop()`, then this signals the
graph is being cancelled early. In this case, all source calculators and graph
input streams will be closed (and remaining Packets will propagate through the
graph).
A source node in a graph will continue to have `Process()` called on it as long
as it returns `::mediapipe::OkStatus(`). To indicate that there is no more data
to be generated return `tool::StatusStop()`. Any other status indicates an error
has occurred.
`Close()` returns `::mediapipe::OkStatus()` to indicate success. Any other
status indicates a failure.
Here is the basic `Process()` function. It uses the `Input()` method (which can
be used only if the calculator has a single input) to request its input data. It
then uses `std::unique_ptr` to allocate the memory needed for the output packet,
and does the calculations. When done it releases the pointer when adding it to
the output stream.
```c++
::util::Status MyCalculator::Process() {
const Matrix& input = Input()->Get<Matrix>();
std::unique_ptr<Matrix> output(new Matrix(input.rows(), input.cols()));
// do your magic here....
// output->row(n) = ...
Output()->Add(output.release(), InputTimestamp());
return ::mediapipe::OkStatus();
}
```
## Example calculator
This section discusses the implementation of `PacketClonerCalculator`, which
does a relatively simple job, and is used in many calculator graphs.
`PacketClonerCalculator` simply produces a copy of its most recent input
packets on demand.
`PacketClonerCalculator` is useful when the timestamps of arriving data packets
are not aligned perfectly. Suppose we have a room with a microphone, light
sensor and a video camera that is collecting sensory data. Each of the sensors
operates independently and collects data intermittently. Suppose that the output
of each sensor is:
* microphone = loudness in decibels of sound in the room (Integer)
* light sensor = brightness of room (Integer)
* video camera = RGB image frame of room (ImageFrame)
Our simple perception pipeline is designed to process sensory data from these 3
sensors such that at any time when we have image frame data from the camera that
is synchronized with the last collected microphone loudness data and light
sensor brightness data. To do this with MediaPipe, our perception pipeline has 3
input streams:
* room_mic_signal - Each packet of data in this input stream is integer data
representing how loud audio is in a room with timestamp.
* room_lightening_sensor - Each packet of data in this input stream is integer
data representing how bright is the room illuminated with timestamp.
* room_video_tick_signal - Each packet of data in this input stream is
imageframe of video data representing video collected from camera in the
room with timestamp.
Below is the implementation of the `PacketClonerCalculator`. You can see
the `GetContract()`, `Open()`, and `Process()` methods as well as the instance
variable `current_` which holds the most recent input packets.
```c++
// This takes packets from N+1 streams, A_1, A_2, ..., A_N, B.
// For every packet that appears in B, outputs the most recent packet from each
// of the A_i on a separate stream.
#include <vector>
#include "absl/strings/str_cat.h"
#include "mediapipe/framework/calculator_framework.h"
namespace mediapipe {
// For every packet received on the last stream, output the latest packet
// obtained on all other streams. Therefore, if the last stream outputs at a
// higher rate than the others, this effectively clones the packets from the
// other streams to match the last.
//
// Example config:
// node {
// calculator: "PacketClonerCalculator"
// input_stream: "first_base_signal"
// input_stream: "second_base_signal"
// input_stream: "tick_signal"
// output_stream: "cloned_first_base_signal"
// output_stream: "cloned_second_base_signal"
// }
//
class PacketClonerCalculator : public CalculatorBase {
public:
static ::mediapipe::Status GetContract(CalculatorContract* cc) {
const int tick_signal_index = cc->Inputs().NumEntries() - 1;
// cc->Inputs().NumEntries() returns the number of input streams
// for the PacketClonerCalculator
for (int i = 0; i < tick_signal_index; ++i) {
cc->Inputs().Index(i).SetAny();
// cc->Inputs().Index(i) returns the input stream pointer by index
cc->Outputs().Index(i).SetSameAs(&cc->Inputs().Index(i));
}
cc->Inputs().Index(tick_signal_index).SetAny();
return ::mediapipe::OkStatus();
}
::mediapipe::Status Open(CalculatorContext* cc) final {
tick_signal_index_ = cc->Inputs().NumEntries() - 1;
current_.resize(tick_signal_index_);
// Pass along the header for each stream if present.
for (int i = 0; i < tick_signal_index_; ++i) {
if (!cc->Inputs().Index(i).Header().IsEmpty()) {
cc->Outputs().Index(i).SetHeader(cc->Inputs().Index(i).Header());
// Sets the output stream of index i header to be the same as
// the header for the input stream of index i
}
}
return ::mediapipe::OkStatus();
}
::mediapipe::Status Process(CalculatorContext* cc) final {
// Store input signals.
for (int i = 0; i < tick_signal_index_; ++i) {
if (!cc->Inputs().Index(i).Value().IsEmpty()) {
current_[i] = cc->Inputs().Index(i).Value();
}
}
// Output if the tick signal is non-empty.
if (!cc->Inputs().Index(tick_signal_index_).Value().IsEmpty()) {
for (int i = 0; i < tick_signal_index_; ++i) {
if (!current_[i].IsEmpty()) {
cc->Outputs().Index(i).AddPacket(
current_[i].At(cc->InputTimestamp()));
// Add a packet to output stream of index i a packet from inputstream i
// with timestamp common to all present inputs
//
} else {
cc->Outputs().Index(i).SetNextTimestampBound(
cc->InputTimestamp().NextAllowedInStream());
// if current_[i], 1 packet buffer for input stream i is empty, we will set
// next allowed timestamp for input stream i to be current timestamp + 1
}
}
}
return ::mediapipe::OkStatus();
}
private:
std::vector<Packet> current_;
int tick_signal_index_;
};
REGISTER_CALCULATOR(PacketClonerCalculator);
} // namespace mediapipe
```
Typically, a calculator has only a .cc file. No .h is required, because
mediapipe uses registration to make calculators known to it. After you have
defined your calculator class, register it with a macro invocation
REGISTER_CALCULATOR(calculator_class_name).
Below is a trivial MediaPipe graph that has 3 input streams, 1 node
(PacketClonerCalculator) and 3 output streams.
```proto
input_stream: "room_mic_signal"
input_stream: "room_lighting_sensor"
input_stream: "room_video_tick_signal"
node {
calculator: "PacketClonerCalculator"
input_stream: "room_mic_signal"
input_stream: "room_lighting_sensor"
input_stream: "room_video_tick_signal"
output_stream: "cloned_room_mic_signal"
output_stream: "cloned_lighting_sensor"
}
```
The diagram below shows how the `PacketClonerCalculator` defines its output
packets based on its series of input packets.
| ![Graph using |
: PacketClonerCalculator](../images/packet_cloner_calculator.png) :
| :--------------------------------------------------------------------------: |
| *Each time it receives a packet on its TICK input stream, the |
: PacketClonerCalculator outputs the most recent packet from each of its input :
: streams. The sequence of output packets is determined by the sequene of :
: input packets and their timestamps. The timestamps are shows along the right :
: side of the diagram.* :

View File

@ -1,24 +1,42 @@
# MediaPipe Concepts
---
layout: default
title: Framework Concepts
nav_order: 5
has_children: true
has_toc: false
---
# Framework Concepts
{: .no_toc }
1. TOC
{:toc}
---
## The basics
### Packet
The basic data flow unit. A packet consists of a numeric timestamp and a shared pointer to an **immutable** payload. The payload can be of any C++ type, and the payload's type is also referred to as the type of the packet. Packets are value classes and can be copied cheaply. Each copy shares ownership of the payload, with reference-counting semantics. Each copy has its own timestamp. [Details](packets.md).
The basic data flow unit. A packet consists of a numeric timestamp and a shared
pointer to an **immutable** payload. The payload can be of any C++ type, and the
payload's type is also referred to as the type of the packet. Packets are value
classes and can be copied cheaply. Each copy shares ownership of the payload,
with reference-counting semantics. Each copy has its own timestamp. See also
[Packet](packets.md).
### Graph
MediaPipe processing takes place inside a graph, which defines packet flow paths
between **nodes**. A graph can have any number of inputs and outputs, and data
flow can branch and merge. Generally data flows forward, but
[backward loops](cycles.md) are possible.
flow can branch and merge. Generally data flows forward, but backward loops are
possible. See [Graphs](graphs.md) for details.
### Nodes
Nodes produce and/or consume packets, and they are where the bulk of the graphs
work takes place. They are also known as “calculators”, for historical reasons.
Each nodes interface defines a number of input and output **ports**, identified by
a tag and/or an index.
Each nodes interface defines a number of input and output **ports**, identified
by a tag and/or an index. See [Calculators](calculators.md) for details.
### Streams
@ -34,21 +52,23 @@ whereas a stream represents a flow of data that changes over time.
### Packet Ports
A port has an associated type; packets transiting through the port must be of
that type. An output stream port can be connected to any number of
input stream ports of the same type; each consumer receives a separate copy of
the output packets, and has its own queue, so it can consume them at its own
pace. Similarly, a side packet output port can be connected to as many side
packet input ports as desired.
that type. An output stream port can be connected to any number of input stream
ports of the same type; each consumer receives a separate copy of the output
packets, and has its own queue, so it can consume them at its own pace.
Similarly, a side packet output port can be connected to as many side packet
input ports as desired.
A port can be required, meaning that a connection must be made for the graph to
be valid, or optional, meaning it may remain unconnected.
Note: even if a stream connection is required, the stream may not carry a packet for all timestamps.
Note: even if a stream connection is required, the stream may not carry a packet
for all timestamps.
## Input and output
Data flow can originate from **source nodes**, which have no input streams and
produce packets spontaneously (e.g. by reading from a file); or from **graph input streams**, which let an application feed packets into a graph.
produce packets spontaneously (e.g. by reading from a file); or from **graph
input streams**, which let an application feed packets into a graph.
Similarly, there are **sink nodes** that receive data and write it to various
destinations (e.g. a file, a memory buffer, etc.), and an application can also
@ -78,15 +98,15 @@ processed data.
### Input policies
The default input policy is deterministic collation of packets by timestamp. A node receives
all inputs for the same timestamp at the same time, in an invocation of its
Process method; and successive input sets are received in their timestamp order. This can
require delaying the processing of some packets until a packet with the same
timestamp is received on all input streams, or until it can be guaranteed that a
packet with that timestamp will not be arriving on the streams that have not
received it.
The default input policy is deterministic collation of packets by timestamp. A
node receives all inputs for the same timestamp at the same time, in an
invocation of its Process method; and successive input sets are received in
their timestamp order. This can require delaying the processing of some packets
until a packet with the same timestamp is received on all input streams, or
until it can be guaranteed that a packet with that timestamp will not be
arriving on the streams that have not received it.
Other policies are also available, implemented using a separate kind of
component known as an InputStreamHandler.
See [scheduling](scheduling_sync.md) for more details.
See [Synchronization](synchronization.md) for more details.

View File

@ -1,14 +1,18 @@
## Running on GPUs
---
layout: default
title: GPU
parent: Framework Concepts
nav_order: 5
---
- [Overview](#overview)
- [OpenGL ES Support](#opengl-es-support)
- [Disable OpenGL ES Support](#disable-opengl-es-support)
- [OpenGL ES Setup on Linux Desktop](#opengl-es-setup-on-linux-desktop)
- [TensorFlow CUDA Support and Setup on Linux Desktop](#tensorflow-cuda-support-and-setup-on-linux-desktop)
- [Life of a GPU Calculator](#life-of-a-gpu-calculator)
- [GpuBuffer to ImageFrame Converters](#gpubuffer-to-imageframe-converters)
# GPU
{: .no_toc }
### Overview
1. TOC
{:toc}
---
## Overview
MediaPipe supports calculator nodes for GPU compute and rendering, and allows combining multiple GPU nodes, as well as mixing them with CPU based calculator nodes. There exist several GPU APIs on mobile platforms (eg, OpenGL ES, Metal and Vulkan). MediaPipe does not attempt to offer a single cross-API GPU abstraction. Individual nodes can be written using different APIs, allowing them to take advantage of platform specific features when needed.
@ -25,7 +29,7 @@ Below are the design principles for GPU support in MediaPipe
* Because different platforms may require different techniques for best performance, the API should allow flexibility in the way things are implemented behind the scenes.
* A calculator should be allowed maximum flexibility in using the GPU for all or part of its operation, combining it with the CPU if necessary.
### OpenGL ES Support
## OpenGL ES Support
MediaPipe supports OpenGL ES up to version 3.2 on Android/Linux and up to ES 3.0
on iOS. In addition, MediaPipe also supports Metal on iOS.
@ -50,172 +54,7 @@ some Android devices. Therefore, our approach is to have one dedicated thread
per context. Each thread issues GL commands, building up a serial command queue
on its context, which is then executed by the GPU asynchronously.
### Disable OpenGL ES Support
By default, building MediaPipe (with no special bazel flags) attempts to compile
and link against OpenGL ES (and for iOS also Metal) libraries.
On platforms where OpenGL ES is not available (see also
[OpenGL ES Setup on Linux Desktop](#opengl-es-setup-on-linux-desktop)), you
should disable OpenGL ES support with:
```
$ bazel build --define MEDIAPIPE_DISABLE_GPU=1 <my-target>
```
Note: On Android and iOS, OpenGL ES is required by MediaPipe framework and the
support should never be disabled.
### OpenGL ES Setup on Linux Desktop
On Linux desktop with video cards that support OpenGL ES 3.1+, MediaPipe can run
GPU compute and rendering and perform TFLite inference on GPU.
To check if your Linux desktop GPU can run MediaPipe with OpenGL ES:
```bash
$ sudo apt-get install mesa-common-dev libegl1-mesa-dev libgles2-mesa-dev
$ sudo apt-get install mesa-utils
$ glxinfo | grep -i opengl
```
For example, it may print:
```bash
$ glxinfo | grep -i opengl
...
OpenGL ES profile version string: OpenGL ES 3.2 NVIDIA 430.50
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20
OpenGL ES profile extensions:
```
*Notice the ES 3.20 text above.*
You need to see ES 3.1 or greater printed in order to perform TFLite inference
on GPU in MediaPipe. With this setup, build with:
```
$ bazel build --copt -DMESA_EGL_NO_X11_HEADERS --copt -DEGL_NO_X11 <my-target>
```
If only ES 3.0 or below is supported, you can still build MediaPipe targets that
don't require TFLite inference on GPU with:
```
$ bazel build --copt -DMESA_EGL_NO_X11_HEADERS --copt -DEGL_NO_X11 --copt -DMEDIAPIPE_DISABLE_GL_COMPUTE <my-target>
```
Note: MEDIAPIPE_DISABLE_GL_COMPUTE is already defined automatically on all Apple
systems (Apple doesn't support OpenGL ES 3.1+).
### TensorFlow CUDA Support and Setup on Linux Desktop
MediaPipe framework doesn't require CUDA for GPU compute and rendering. However,
MediaPipe can work with TensorFlow to perform GPU inference on video cards that
support CUDA.
To enable TensorFlow GPU inference with MediaPipe, the first step is to follow
the
[TensorFlow GPU documentation](https://www.tensorflow.org/install/gpu#software_requirements)
to install the required NVIDIA software on your Linux desktop.
After installation, update `$PATH` and `$LD_LIBRARY_PATH` and run `ldconfig`
with:
```
$ export PATH=/usr/local/cuda-10.1/bin${PATH:+:${PATH}}
$ export LD_LIBRARY_PATH=/usr/local/cuda/extras/CUPTI/lib64,/usr/local/cuda-10.1/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
$ sudo ldconfig
```
It's recommended to verify the installation of CUPTI, CUDA, CuDNN, and NVCC:
```
$ 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
$ ls /usr/local/cuda-10.1
LICENSE bin extras lib64 libnvvp nvml samples src tools
README doc include libnsight nsightee_plugins nvvm share targets version.txt
$ 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
$ ls /usr/lib/x86_64-linux-gnu/ | grep libcudnn.so
libcudnn.so
libcudnn.so.7
libcudnn.so.7.6.4
```
Setting `$TF_CUDA_PATHS` is the way to declare where the CUDA library is. Note
that the following code snippet also adds `/usr/lib/x86_64-linux-gnu` and
`/usr/include` into `$TF_CUDA_PATHS` for cudablas and libcudnn.
```
$ export TF_CUDA_PATHS=/usr/local/cuda-10.1,/usr/lib/x86_64-linux-gnu,/usr/include
```
To make MediaPipe get TensorFlow's CUDA settings, find TensorFlow's
[.bazelrc](https://github.com/tensorflow/tensorflow/blob/master/.bazelrc) and
copy the `build:using_cuda` and `build:cuda` section into MediaPipe's .bazelrc
file. For example, as of April 23, 2020, TensorFlow's CUDA setting is the
following:
```
# This config refers to building with CUDA available. It does not necessarily
# mean that we build CUDA op kernels.
build:using_cuda --define=using_cuda=true
build:using_cuda --action_env TF_NEED_CUDA=1
build:using_cuda --crosstool_top=@local_config_cuda//crosstool:toolchain
# This config refers to building CUDA op kernels with nvcc.
build:cuda --config=using_cuda
build:cuda --define=using_cuda_nvcc=true
```
Finally, build MediaPipe with TensorFlow GPU with two more flags `--config=cuda`
and `--spawn_strategy=local`. For example:
```
$ bazel build -c opt --config=cuda --spawn_strategy=local \
--define no_aws_support=true --copt -DMESA_EGL_NO_X11_HEADERS \
mediapipe/examples/desktop/object_detection:object_detection_tensorflow
```
While the binary is running, it prints out the GPU device info:
```
I external/org_tensorflow/tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
I external/org_tensorflow/tensorflow/core/common_runtime/gpu/gpu_device.cc:1544] Found device 0 with properties: pciBusID: 0000:00:04.0 name: Tesla T4 computeCapability: 7.5 coreClock: 1.59GHz coreCount: 40 deviceMemorySize: 14.75GiB deviceMemoryBandwidth: 298.08GiB/s
I external/org_tensorflow/tensorflow/core/common_runtime/gpu/gpu_device.cc:1686] Adding visible gpu devices: 0
```
You can monitor the GPU usage to verify whether the GPU is used for model
inference.
```
$ nvidia-smi --query-gpu=utilization.gpu --format=csv --loop=1
0 %
0 %
4 %
5 %
83 %
21 %
22 %
27 %
29 %
100 %
0 %
0%
```
### Life of a GPU Calculator
## Life of a GPU Calculator
This section presents the basic structure of the Process method of a GPU
calculator derived from base class GlSimpleCalculator. The GPU calculator
@ -302,7 +141,7 @@ choices for MediaPipe GPU support:
* Data that needs to be shared between all GPU-based calculators is provided as a external input that is implemented as a graph service and is managed by the `GlCalculatorHelper` class.
* The combination of calculator-specific helpers and a shared graph service allows us great flexibility in managing the GPU resource: we can have a separate context per calculator, share a single context, share a lock or other synchronization primitives, etc. -- and all of this is managed by the helper and hidden from the individual calculators.
### GpuBuffer to ImageFrame Converters
## GpuBuffer to ImageFrame Converters
We provide two calculators called `GpuBufferToImageFrameCalculator` and `ImageFrameToGpuBufferCalculator`. These calculators convert between `ImageFrame` and `GpuBuffer`, allowing the construction of graphs that combine GPU and CPU calculators. They are supported on both iOS and Android
@ -310,6 +149,15 @@ When possible, these calculators use platform-specific functionality to share da
The below diagram shows the data flow in a mobile application that captures video from the camera, runs it through a MediaPipe graph, and renders the output on the screen in real time. The dashed line indicates which parts are inside the MediaPipe graph proper. This application runs a Canny edge-detection filter on the CPU using OpenCV, and overlays it on top of the original video using the GPU.
| ![How GPU calculators interact](images/gpu_example_graph.png) |
|:--:|
| *Video frames from the camera are fed into the graph as `GpuBuffer` packets. The input stream is accessed by two calculators in parallel. `GpuBufferToImageFrameCalculator` converts the buffer into an `ImageFrame`, which is then sent through a grayscale converter and a canny filter (both based on OpenCV and running on the CPU), whose output is then converted into a `GpuBuffer` again. A multi-input GPU calculator, GlOverlayCalculator, takes as input both the original `GpuBuffer` and the one coming out of the edge detector, and overlays them using a shader. The output is then sent back to the application using a callback calculator, and the application renders the image to the screen using OpenGL.* |
| ![How GPU calculators interact](../images/gpu_example_graph.png) |
| :--------------------------------------------------------------------------: |
| *Video frames from the camera are fed into the graph as `GpuBuffer` packets. |
: The input stream is accessed by two calculators in parallel. :
: `GpuBufferToImageFrameCalculator` converts the buffer into an `ImageFrame`, :
: which is then sent through a grayscale converter and a canny filter (both :
: based on OpenCV and running on the CPU), whose output is then converted into :
: a `GpuBuffer` again. A multi-input GPU calculator, GlOverlayCalculator, :
: takes as input both the original `GpuBuffer` and the one coming out of the :
: edge detector, and overlays them using a shader. The output is then sent :
: back to the application using a callback calculator, and the application :
: renders the image to the screen using OpenGL.* :

View File

@ -0,0 +1,271 @@
---
layout: default
title: Graphs
parent: Framework Concepts
nav_order: 2
---
# Graphs
{: .no_toc }
1. TOC
{:toc}
---
## GraphConfig
A `GraphConfig` is a specification that describes the topology and functionality
of a MediaPipe graph. In the specification, a node in the graph represents an
instance of a particular calculator. All the necessary configurations of the
node, such its type, inputs and outputs must be described in the specification.
Description of the node can also include several optional fields, such as
node-specific options, input policy and executor, discussed in
[Synchronization](synchronization.md).
`GraphConfig` has several other fields to configure the global graph-level
settings, eg, graph executor configs, number of threads, and maximum queue size
of input streams. Several graph-level settings are useful for tuning the
performance of the graph on different platforms (eg, desktop v.s. mobile). For
instance, on mobile, attaching a heavy model-inference calculator to a separate
executor can improve the performance of a real-time application since this
enables thread locality.
Below is a trivial `GraphConfig` example where we have series of passthrough
calculators :
```proto
# This graph named main_pass_throughcals_nosubgraph.pbtxt contains 4
# passthrough calculators.
input_stream: "in"
node {
calculator: "PassThroughCalculator"
input_stream: "in"
output_stream: "out1"
}
node {
calculator: "PassThroughCalculator"
input_stream: "out1"
output_stream: "out2"
}
node {
calculator: "PassThroughCalculator"
input_stream: "out2"
output_stream: "out3"
}
node {
calculator: "PassThroughCalculator"
input_stream: "out3"
output_stream: "out4"
}
```
## Subgraph
To modularize a `CalculatorGraphConfig` into sub-modules and assist with re-use
of perception solutions, a MediaPipe graph can be defined as a `Subgraph`. The
public interface of a subgraph consists of a set of input and output streams
similar to a calculator's public interface. The subgraph can then be included in
an `CalculatorGraphConfig` as if it were a calculator. When a MediaPipe graph is
loaded from a `CalculatorGraphConfig`, each subgraph node is replaced by the
corresponding graph of calculators. As a result, the semantics and performance
of the subgraph is identical to the corresponding graph of calculators.
Below is an example of how to create a subgraph named `TwoPassThroughSubgraph`.
1. Defining the subgraph.
```proto
# This subgraph is defined in two_pass_through_subgraph.pbtxt
# and is registered as "TwoPassThroughSubgraph"
type: "TwoPassThroughSubgraph"
input_stream: "out1"
output_stream: "out3"
node {
calculator: "PassThroughculator"
input_stream: "out1"
output_stream: "out2"
}
node {
calculator: "PassThroughculator"
input_stream: "out2"
output_stream: "out3"
}
```
The public interface to the subgraph consists of:
* Graph input streams
* Graph output streams
* Graph input side packets
* Graph output side packets
2. Register the subgraph using BUILD rule `mediapipe_simple_subgraph`. The
parameter `register_as` defines the component name for the new subgraph.
```proto
# Small section of BUILD file for registering the "TwoPassThroughSubgraph"
# subgraph for use by main graph main_pass_throughcals.pbtxt
mediapipe_simple_subgraph(
name = "twopassthrough_subgraph",
graph = "twopassthrough_subgraph.pbtxt",
register_as = "TwoPassThroughSubgraph",
deps = [
"//mediapipe/calculators/core:pass_through_calculator",
"//mediapipe/framework:calculator_graph",
],
)
```
3. Use the subgraph in the main graph.
```proto
# This main graph is defined in main_pass_throughcals.pbtxt
# using subgraph called "TwoPassThroughSubgraph"
input_stream: "in"
node {
calculator: "PassThroughCalculator"
input_stream: "in"
output_stream: "out1"
}
node {
calculator: "TwoPassThroughSubgraph"
input_stream: "out1"
output_stream: "out3"
}
node {
calculator: "PassThroughCalculator"
input_stream: "out3"
output_stream: "out4"
}
```
## Cycles
<!-- TODO: add discussion of PreviousLoopbackCalculator -->
By default, MediaPipe requires calculator graphs to be acyclic and treats cycles
in a graph as errors. If a graph is intended to have cycles, the cycles need to
be annotated in the graph config. This page describes how to do that.
NOTE: The current approach is experimental and subject to change. We welcome
your feedback.
Please use the `CalculatorGraphTest.Cycle` unit test in
`mediapipe/framework/calculator_graph_test.cc` as sample code. Shown
below is the cyclic graph in the test. The `sum` output of the adder is the sum
of the integers generated by the integer source calculator.
![a cyclic graph that adds a stream of integers](../images/cyclic_integer_sum_graph.svg "A cyclic graph")
This simple graph illustrates all the issues in supporting cyclic graphs.
### Back Edge Annotation
We require that an edge in each cycle be annotated as a back edge. This allows
MediaPipes topological sort to work, after removing all the back edges.
There are usually multiple ways to select the back edges. Which edges are marked
as back edges affects which nodes are considered as upstream and which nodes are
considered as downstream, which in turn affects the priorities MediaPipe assigns
to the nodes.
For example, the `CalculatorGraphTest.Cycle` test marks the `old_sum` edge as a
back edge, so the Delay node is considered as a downstream node of the adder
node and is given a higher priority. Alternatively, we could mark the `sum`
input to the delay node as the back edge, in which case the delay node would be
considered as an upstream node of the adder node and is given a lower priority.
### Initial Packet
For the adder calculator to be runnable when the first integer from the integer
source arrives, we need an initial packet, with value 0 and with the same
timestamp, on the `old_sum` input stream to the adder. This initial packet
should be output by the delay calculator in the `Open()` method.
### Delay in a Loop
Each loop should incur a delay to align the previous `sum` output with the next
integer input. This is also done by the delay node. So the delay node needs to
know the following about the timestamps of the integer source calculator:
* The timestamp of the first output.
* The timestamp delta between successive outputs.
We plan to add an alternative scheduling policy that only cares about packet
ordering and ignores packet timestamps, which will eliminate this inconvenience.
### Early Termination of a Calculator When One Input Stream is Done
By default, MediaPipe calls the `Close()` method of a non-source calculator when
all of its input streams are done. In the example graph, we want to stop the
adder node as soon as the integer source is done. This is accomplished by
configuring the adder node with an alternative input stream handler,
`EarlyCloseInputStreamHandler`.
### Relevant Source Code
#### Delay Calculator
Note the code in `Open()` that outputs the initial packet and the code in
`Process()` that adds a (unit) delay to input packets. As noted above, this
delay node assumes that its output stream is used alongside an input stream with
packet timestamps 0, 1, 2, 3, ...
```c++
class UnitDelayCalculator : public Calculator {
public:
 static ::util::Status FillExpectations(
     const CalculatorOptions& extendable_options, PacketTypeSet* inputs,
     PacketTypeSet* outputs, PacketTypeSet* input_side_packets) {
   inputs->Index(0)->Set<int>("An integer.");
   outputs->Index(0)->Set<int>("The input delayed by one time unit.");
   return ::mediapipe::OkStatus();
 }
 ::util::Status Open() final {
   Output()->Add(new int(0), Timestamp(0));
   return ::mediapipe::OkStatus();
 }
 ::util::Status Process() final {
   const Packet& packet = Input()->Value();
   Output()->AddPacket(packet.At(packet.Timestamp().NextAllowedInStream()));
   return ::mediapipe::OkStatus();
 }
};
```
#### Graph Config
Note the `back_edge` annotation and the alternative `input_stream_handler`.
```proto
node {
  calculator: 'GlobalCountSourceCalculator'
  input_side_packet: 'global_counter'
  output_stream: 'integers'
}
node {
  calculator: 'IntAdderCalculator'
  input_stream: 'integers'
  input_stream: 'old_sum'
  input_stream_info: {
    tag_index: ':1' # 'old_sum'
    back_edge: true
  }
  output_stream: 'sum'
  input_stream_handler {
    input_stream_handler: 'EarlyCloseInputStreamHandler'
  }
}
node {
  calculator: 'UnitDelayCalculator'
  input_stream: 'sum'
  output_stream: 'old_sum'
}
```

View File

@ -1,10 +1,21 @@
### Packets
---
layout: default
title: Packets
parent: Framework Concepts
nav_order: 3
---
- [Creating a packet](#creating-a-packet)
# Packets
{: .no_toc }
1. TOC
{:toc}
---
Each calculator is a node of of a graph. We describe how to create a new calculator, how to initialize a calculator, how to perform its calculations, input and output streams, timestamps, and options
#### Creating a packet
## Creating a packet
Packets are generally created with `MediaPipe::Adopt()` (from packet.h).
```c++

View File

@ -1,4 +1,16 @@
# Framework Architecture
---
layout: default
title: Synchronization
parent: Framework Concepts
nav_order: 4
---
# Synchronization
{: .no_toc }
1. TOC
{:toc}
---
## Scheduling mechanics

View File

@ -1,4 +1,16 @@
## MediaPipe Android Archive Library
---
layout: default
title: MediaPipe Android Archive
parent: Getting Started
nav_order: 7
---
# MediaPipe Android Archive
{: .no_toc }
1. TOC
{:toc}
---
***Experimental Only***
@ -9,15 +21,15 @@ target to generate a custom AAR file for their own projects. This is necessary
in order to include specific resources such as MediaPipe calculators needed for
each project.
### Steps to build a MediaPipe AAR
## Steps to build a MediaPipe AAR
1. Create a mediapipe_aar() target.
In the MediaPipe directory, create a new mediapipe_aar() target in a BUILD
file. You need to figure out what calculators are used in the graph and
provide the calculator dependencies to the mediapipe_aar(). For example, to
build an AAR for [face detection gpu](./face_detection_mobile_gpu.md), you
can put the following code into
build an AAR for [MediaPipe Face Detection](../solutions/face_detection.md),
you can put the following code into
mediapipe/examples/android/src/java/com/google/mediapipe/apps/aar_example/BUILD.
```
@ -54,7 +66,7 @@ each project.
/absolute/path/to/your/preferred/location
```
### Steps to use a MediaPipe AAR in Android Studio with Gradle
## Steps to use a MediaPipe AAR in Android Studio with Gradle
1. Start Android Studio and go to your project.
@ -65,7 +77,7 @@ each project.
/path/to/your/app/libs/
```
![Screenshot](images/mobile/aar_location.png)
![Screenshot](../images/mobile/aar_location.png)
3. Make app/src/main/assets and copy assets (graph, model, and etc) into
app/src/main/assets.
@ -85,7 +97,7 @@ each project.
cp mediapipe/models/face_detection_front_labelmap.txt /path/to/your/app/src/main/assets/
```
![Screenshot](images/mobile/assets_location.png)
![Screenshot](../images/mobile/assets_location.png)
4. Make app/src/main/jniLibs and copy OpenCV JNI libraries into
app/src/main/jniLibs.
@ -100,7 +112,7 @@ each project.
cp -R ~/Downloads/OpenCV-android-sdk/sdk/native/libs/arm* /path/to/your/app/src/main/jniLibs/
```
![Screenshot](images/mobile/android_studio_opencv_location.png)
![Screenshot](../images/mobile/android_studio_opencv_location.png)
5. Modify app/build.gradle to add MediaPipe dependencies and MediaPipe AAR.
@ -127,6 +139,8 @@ each project.
```
6. Follow our Android app examples to use MediaPipe in Android Studio for your
use case. If you are looking for an example, a face detection
example can be found
[here](https://github.com/jiuqiant/mediapipe_face_detection_aar_example) and a multi-hand tracking example can be found [here](https://github.com/jiuqiant/mediapipe_multi_hands_tracking_aar_example).
use case. If you are looking for an example, a face detection example can be
found
[here](https://github.com/jiuqiant/mediapipe_face_detection_aar_example) and
a multi-hand tracking example can be found
[here](https://github.com/jiuqiant/mediapipe_multi_hands_tracking_aar_example).

View File

@ -1,8 +1,16 @@
# Building MediaPipe Examples
---
layout: default
title: Building MediaPipe Examples
parent: Getting Started
nav_order: 2
---
* [Android](#android)
* [iOS](#ios)
* [Desktop](#desktop)
# Building MediaPipe Examples
{: .no_toc }
1. TOC
{:toc}
---
## Android
@ -29,8 +37,9 @@ export ANDROID_NDK_HOME=<path to the Android NDK>
In order to use MediaPipe on earlier Android versions, MediaPipe needs to switch
to a lower Android API level. You can achieve this by specifying `api_level =
<api level integer>` in android_ndk_repository() and/or android_sdk_repository()
in the [`WORKSPACE`](https://github.com/google/mediapipe/tree/master/WORKSPACE) file.
$YOUR_INTENDED_API_LEVEL` in android_ndk_repository() and/or
android_sdk_repository() in the
[`WORKSPACE`](https://github.com/google/mediapipe/tree/master/WORKSPACE) file.
Please verify all the necessary packages are installed.
@ -42,16 +51,19 @@ Please verify all the necessary packages are installed.
### Option 1: Build with Bazel in Command Line
1. To build an Android example app, for instance, for MediaPipe Hand, run:
1. To build an Android example app, build against the corresponding
`android_binary` build target. For instance, for
[MediaPipe Hand](../solutions/hand.md) the target is `handtrackinggpu` in
the
[BUILD](https://github.com/google/mediapipe/tree/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/handtrackinggpu/BUILD)
file:
Note: To reduce the binary size, consider appending `--linkopt="-s"` to the
command below to strip symbols.
~~~
```bash
bazel build -c opt --config=android_arm64 mediapipe/examples/android/src/java/com/google/mediapipe/apps/handtrackinggpu
bazel build -c opt --config=android_arm64 mediapipe/examples/android/src/java/com/google/mediapipe/apps/handtrackinggpu:handtrackinggpu
```
~~~
1. Install it on a device with:
@ -74,13 +86,13 @@ app:
1. Install and launch Android Studio 3.5.
2. Select `Configure` | `SDK Manager` | `SDK Platforms`.
2. Select `Configure` -> `SDK Manager` -> `SDK Platforms`.
* Verify that Android SDK Platform API Level 28 or 29 is installed.
* Take note of the Android SDK Location, e.g.,
`/usr/local/home/Android/Sdk`.
3. Select `Configure` | `SDK Manager` | `SDK Tools`.
3. Select `Configure` -> `SDK Manager` -> `SDK Tools`.
* Verify that Android SDK Build-Tools 28 or 29 is installed.
* Verify that Android SDK Platform-Tools 28 or 29 is installed.
@ -102,10 +114,10 @@ app:
export ANDROID_NDK_HOME=/usr/local/home/Android/Sdk/ndk/<version number>
```
5. Select `Configure` | `Plugins` install `Bazel`.
5. Select `Configure` -> `Plugins` to install `Bazel`.
6. On Linux, select `File` | `Settings`| `Bazel settings`. On macos, select
`Android Studio` | `Preferences` | `Bazel settings`. Then, modify `Bazel
6. On Linux, select `File` -> `Settings` -> `Bazel settings`. On macos, select
`Android Studio` -> `Preferences` -> `Bazel settings`. Then, modify `Bazel
binary location` to be the same as the output of `$ which bazel`.
7. Select `Import Bazel Project`.
@ -132,7 +144,7 @@ app:
--host_crosstool_top=@bazel_tools//tools/cpp:toolchain
```
8. Select `Bazel` | `Sync` | `Sync project with Build files`.
8. Select `Bazel` -> `Sync` -> `Sync project with Build files`.
Note: Even after doing step 4, if you still see the error: `"no such package
'@androidsdk//': Either the path attribute of android_sdk_repository or the
@ -154,11 +166,11 @@ app:
9. Connect an Android device to the workstation.
10. Select `Run...` | `Edit Configurations...`.
10. Select `Run...` -> `Edit Configurations...`.
* Select `Templates` | `Bazel Command`.
* Select `Templates` -> `Bazel Command`.
* Enter Target Expression:
`//mediapipe/examples/android/src/java/com/google/mediapipe/apps/facedetectioncpu`
`//mediapipe/examples/android/src/java/com/google/mediapipe/apps/handtrackinggpu:handtrackinggpu`
* Enter Bazel command: `mobile-install`.
* Enter Bazel flags: `-c opt --config=android_arm64`.
* Press the `[+]` button to add the new configuration.
@ -215,15 +227,14 @@ app:
### Option 1: Build with Bazel in Command Line
1. Modify the `bundle_id` field of the app's `ios_application` target to use
your own identifier. For instance, for
[MediaPipe Hand](./hand_tracking_mobile_gpu.md), the `bundle_id` is in the
1. Modify the `bundle_id` field of the app's `ios_application` build target to
use your own identifier. For instance, for
[MediaPipe Hand](../solutions/hand.md), the `bundle_id` is in the
`HandTrackingGpuApp` target in the
[BUILD](https://github.com/google/mediapipe/tree/master/mediapipe/examples/ios/handtrackinggpu/BUILD)
file.
2. Again using [MediaPipe Hand](./hand_tracking_mobile_gpu.md) for example,
run:
2. Again using [MediaPipe Hand](../solutions/hand.md) for example, run:
```bash
bazel build -c opt --config=ios_arm64 mediapipe/examples/ios/handtrackinggpu:HandTrackingGpuApp
@ -287,7 +298,7 @@ the previous section.
### Option 1: Running on CPU
1. To build, for example, [MediaPipe Hand](./hand_tracking_mobile_gpu.md), run:
1. To build, for example, [MediaPipe Hand](../solutions/hand.md), run:
```bash
bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/examples/desktop/hand_tracking:hand_tracking_cpu
@ -306,9 +317,9 @@ the previous section.
### Option 2: Running on GPU
Note: This currently works only on Linux, and please first follow
[OpenGL ES Setup on Linux Desktop](./gpu.md#opengl-es-setup-on-linux-desktop).
[OpenGL ES Setup on Linux Desktop](./gpu_support.md#opengl-es-setup-on-linux-desktop).
1. To build, for example, [MediaPipe Hand](./hand_tracking_mobile_gpu.md), run:
1. To build, for example, [MediaPipe Hand](../solutions/hand.md), run:
```bash
bazel build -c opt --copt -DMESA_EGL_NO_X11_HEADERS --copt -DEGL_NO_X11 \

View File

@ -1,12 +1,16 @@
## Questions and Answers
---
layout: default
title: FAQ
parent: Getting Started
nav_order: 9
---
- [How to convert ImageFrames and GpuBuffers](#how-to-convert-imageframes-and-gpubuffers)
- [How to visualize perceived results](#how-to-visualize-perception-results)
- [How to run calculators in parallel](#how-to-run-calculators-in-parallel)
- [Output timestamps when using ImmediateInputStreamHandler](#output-timestamps-when-using-immediateinputstreamhandler)
- [How to change settings at runtime](#how-to-change-settings-at-runtime)
- [How to process real-time input streams](#how-to-process-real-time-input-streams)
- [Can I run MediaPipe on MS Windows?](#can-i-run-mediapipe-on-ms-windows)
# FAQ
{: .no_toc }
1. TOC
{:toc}
---
### How to convert ImageFrames and GpuBuffers

View File

@ -0,0 +1,13 @@
---
layout: default
title: Getting Started
nav_order: 2
has_children: true
---
# Getting Started
{: .no_toc }
1. TOC
{:toc}
---

View File

@ -0,0 +1,186 @@
---
layout: default
title: GPU Support
parent: Getting Started
nav_order: 6
---
# GPU Support
{: .no_toc }
1. TOC
{:toc}
---
## OpenGL ES Support
MediaPipe supports OpenGL ES up to version 3.2 on Android/Linux and up to ES 3.0
on iOS. In addition, MediaPipe also supports Metal on iOS.
OpenGL ES 3.1 or greater is required (on Android/Linux systems) for running
machine learning inference calculators and graphs.
## Disable OpenGL ES Support
By default, building MediaPipe (with no special bazel flags) attempts to compile
and link against OpenGL ES (and for iOS also Metal) libraries.
On platforms where OpenGL ES is not available (see also
[OpenGL ES Setup on Linux Desktop](#opengl-es-setup-on-linux-desktop)), you
should disable OpenGL ES support with:
```
$ bazel build --define MEDIAPIPE_DISABLE_GPU=1 <my-target>
```
Note: On Android and iOS, OpenGL ES is required by MediaPipe framework and the
support should never be disabled.
## OpenGL ES Setup on Linux Desktop
On Linux desktop with video cards that support OpenGL ES 3.1+, MediaPipe can run
GPU compute and rendering and perform TFLite inference on GPU.
To check if your Linux desktop GPU can run MediaPipe with OpenGL ES:
```bash
$ sudo apt-get install mesa-common-dev libegl1-mesa-dev libgles2-mesa-dev
$ sudo apt-get install mesa-utils
$ glxinfo | grep -i opengl
```
For example, it may print:
```bash
$ glxinfo | grep -i opengl
...
OpenGL ES profile version string: OpenGL ES 3.2 NVIDIA 430.50
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20
OpenGL ES profile extensions:
```
*Notice the ES 3.20 text above.*
You need to see ES 3.1 or greater printed in order to perform TFLite inference
on GPU in MediaPipe. With this setup, build with:
```
$ bazel build --copt -DMESA_EGL_NO_X11_HEADERS --copt -DEGL_NO_X11 <my-target>
```
If only ES 3.0 or below is supported, you can still build MediaPipe targets that
don't require TFLite inference on GPU with:
```
$ bazel build --copt -DMESA_EGL_NO_X11_HEADERS --copt -DEGL_NO_X11 --copt -DMEDIAPIPE_DISABLE_GL_COMPUTE <my-target>
```
Note: MEDIAPIPE_DISABLE_GL_COMPUTE is already defined automatically on all Apple
systems (Apple doesn't support OpenGL ES 3.1+).
## TensorFlow CUDA Support and Setup on Linux Desktop
MediaPipe framework doesn't require CUDA for GPU compute and rendering. However,
MediaPipe can work with TensorFlow to perform GPU inference on video cards that
support CUDA.
To enable TensorFlow GPU inference with MediaPipe, the first step is to follow
the
[TensorFlow GPU documentation](https://www.tensorflow.org/install/gpu#software_requirements)
to install the required NVIDIA software on your Linux desktop.
After installation, update `$PATH` and `$LD_LIBRARY_PATH` and run `ldconfig`
with:
```
$ export PATH=/usr/local/cuda-10.1/bin${PATH:+:${PATH}}
$ export LD_LIBRARY_PATH=/usr/local/cuda/extras/CUPTI/lib64,/usr/local/cuda-10.1/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
$ sudo ldconfig
```
It's recommended to verify the installation of CUPTI, CUDA, CuDNN, and NVCC:
```
$ 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
$ ls /usr/local/cuda-10.1
LICENSE bin extras lib64 libnvvp nvml samples src tools
README doc include libnsight nsightee_plugins nvvm share targets version.txt
$ 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
$ ls /usr/lib/x86_64-linux-gnu/ | grep libcudnn.so
libcudnn.so
libcudnn.so.7
libcudnn.so.7.6.4
```
Setting `$TF_CUDA_PATHS` is the way to declare where the CUDA library is. Note
that the following code snippet also adds `/usr/lib/x86_64-linux-gnu` and
`/usr/include` into `$TF_CUDA_PATHS` for cudablas and libcudnn.
```
$ export TF_CUDA_PATHS=/usr/local/cuda-10.1,/usr/lib/x86_64-linux-gnu,/usr/include
```
To make MediaPipe get TensorFlow's CUDA settings, find TensorFlow's
[.bazelrc](https://github.com/tensorflow/tensorflow/blob/master/.bazelrc) and
copy the `build:using_cuda` and `build:cuda` section into MediaPipe's .bazelrc
file. For example, as of April 23, 2020, TensorFlow's CUDA setting is the
following:
```
# This config refers to building with CUDA available. It does not necessarily
# mean that we build CUDA op kernels.
build:using_cuda --define=using_cuda=true
build:using_cuda --action_env TF_NEED_CUDA=1
build:using_cuda --crosstool_top=@local_config_cuda//crosstool:toolchain
# This config refers to building CUDA op kernels with nvcc.
build:cuda --config=using_cuda
build:cuda --define=using_cuda_nvcc=true
```
Finally, build MediaPipe with TensorFlow GPU with two more flags `--config=cuda`
and `--spawn_strategy=local`. For example:
```
$ bazel build -c opt --config=cuda --spawn_strategy=local \
--define no_aws_support=true --copt -DMESA_EGL_NO_X11_HEADERS \
mediapipe/examples/desktop/object_detection:object_detection_tensorflow
```
While the binary is running, it prints out the GPU device info:
```
I external/org_tensorflow/tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
I external/org_tensorflow/tensorflow/core/common_runtime/gpu/gpu_device.cc:1544] Found device 0 with properties: pciBusID: 0000:00:04.0 name: Tesla T4 computeCapability: 7.5 coreClock: 1.59GHz coreCount: 40 deviceMemorySize: 14.75GiB deviceMemoryBandwidth: 298.08GiB/s
I external/org_tensorflow/tensorflow/core/common_runtime/gpu/gpu_device.cc:1686] Adding visible gpu devices: 0
```
You can monitor the GPU usage to verify whether the GPU is used for model
inference.
```
$ nvidia-smi --query-gpu=utilization.gpu --format=csv --loop=1
0 %
0 %
4 %
5 %
83 %
21 %
22 %
27 %
29 %
100 %
0 %
0%
```

View File

@ -1,4 +1,16 @@
# Hello World! in MediaPipe on Android
---
layout: default
title: Hello World! on Android
parent: Getting Started
nav_order: 3
---
# Hello World! on Android
{: .no_toc }
1. TOC
{:toc}
---
## Introduction
@ -14,7 +26,7 @@ graph on Android.
A simple camera app for real-time Sobel edge detection applied to a live video
stream on an Android device.
![edge_detection_android_gpu_gif](images/mobile/edge_detection_android_gpu.gif)
![edge_detection_android_gpu_gif](../images/mobile/edge_detection_android_gpu.gif)
## Setup
@ -56,7 +68,7 @@ node: {
A visualization of the graph is shown below:
![edge_detection_mobile_gpu](images/mobile/edge_detection_mobile_gpu.png)
![edge_detection_mobile_gpu](../images/mobile/edge_detection_mobile_gpu.png)
This graph has a single input stream named `input_video` for all incoming frames
that will be provided by your device's camera.
@ -247,7 +259,7 @@ adb install bazel-bin/$APPLICATION_PATH/helloworld.apk
Open the application on your device. It should display a screen with the text
`Hello World!`.
![bazel_hello_world_android](images/mobile/bazel_hello_world_android.png)
![bazel_hello_world_android](../images/mobile/bazel_hello_world_android.png)
## Using the camera via `CameraX`
@ -364,7 +376,7 @@ Add the following line in the `$APPLICATION_PATH/res/values/strings.xml` file:
When the user doesn't grant camera permission, the screen will now look like
this:
![missing_camera_permission_android](images/mobile/missing_camera_permission_android.png)
![missing_camera_permission_android](../images/mobile/missing_camera_permission_android.png)
Now, we will add the [`SurfaceTexture`] and [`SurfaceView`] objects to
`MainActivity`:
@ -740,7 +752,7 @@ And that's it! You should now be able to successfully build and run the
application on the device and see Sobel edge detection running on a live camera
feed! Congrats!
![edge_detection_android_gpu_gif](images/mobile/edge_detection_android_gpu.gif)
![edge_detection_android_gpu_gif](../images/mobile/edge_detection_android_gpu.gif)
If you ran into any issues, please see the full code of the tutorial
[here](https://github.com/google/mediapipe/tree/master/mediapipe/examples/android/src/java/com/google/mediapipe/apps/basic).

View File

@ -1,4 +1,16 @@
## Hello World for C++
---
layout: default
title: Hello World! on Desktop (C++)
parent: Getting Started
nav_order: 5
---
# Hello World! on Desktop (C++)
{: .no_toc }
1. TOC
{:toc}
---
1. Ensure you have a working version of MediaPipe. See
[installation instructions](./install.md).
@ -52,7 +64,7 @@
You can visualize this graph using
[MediaPipe Visualizer](https://viz.mediapipe.dev) by pasting the
CalculatorGraphConfig content below into the visualizer. See
[here](./visualizer.md) for help on the visualizer.
[here](../tools/visualizer.md) for help on the visualizer.
```bash
input_stream: "in"
@ -72,7 +84,7 @@
This graph consists of 1 graph input stream (`in`) and 1 graph output stream
(`out`), and 2 [`PassThroughCalculator`]s connected serially.
![hello_world graph](./images/hello_world.png)
![hello_world graph](../images/hello_world.png)
4. Before running the graph, an `OutputStreamPoller` object is connected to the
output stream in order to later retrieve the graph output, and a graph run

View File

@ -1,4 +1,16 @@
# Hello World! in MediaPipe on iOS
---
layout: default
title: Hello World! on iOS
parent: Getting Started
nav_order: 4
---
# Hello World! on iOS
{: .no_toc }
1. TOC
{:toc}
---
## Introduction
@ -14,7 +26,7 @@ graph on iOS.
A simple camera app for real-time Sobel edge detection applied to a live video
stream on an iOS device.
![edge_detection_ios_gpu_gif](images/mobile/edge_detection_ios_gpu.gif)
![edge_detection_ios_gpu_gif](../images/mobile/edge_detection_ios_gpu.gif)
## Setup
@ -54,7 +66,7 @@ node: {
A visualization of the graph is shown below:
![edge_detection_mobile_gpu](images/mobile/edge_detection_mobile_gpu.png)
![edge_detection_mobile_gpu](../images/mobile/edge_detection_mobile_gpu.png)
This graph has a single input stream named `input_video` for all incoming frames
that will be provided by your device's camera.
@ -538,7 +550,7 @@ method to receive packets on this output stream and display them on the screen:
And that is all! Build and run the app on your iOS device. You should see the
results of running the edge detection graph on a live video feed. Congrats!
![edge_detection_ios_gpu_gif](images/mobile/edge_detection_ios_gpu.gif)
![edge_detection_ios_gpu_gif](../images/mobile/edge_detection_ios_gpu.gif)
If you ran into any issues, please see the full code of the tutorial
[here](https://github.com/google/mediapipe/tree/master/mediapipe/examples/ios/edgedetectiongpu).

View File

@ -1,17 +1,24 @@
## Getting Help
---
layout: default
title: Getting Help
parent: Getting Started
nav_order: 8
---
- [Technical questions](#technical-questions)
- [Bugs and feature requests](#bugs-and-feature-requests)
# Getting Help
{: .no_toc }
Below are the various ways to get help:
1. TOC
{:toc}
---
### Technical questions
## Technical questions
For help with technical or algorithmic questions, visit
[Stack Overflow](https://stackoverflow.com/questions/tagged/mediapipe) to find
answers and support from the MediaPipe community.
### Bugs and feature requests
## Bugs and feature requests
To report bugs or make feature requests,
[file an issue on GitHub](https://github.com/google/mediapipe/issues).

View File

@ -1,4 +1,16 @@
## Installing MediaPipe
---
layout: default
title: Installation
parent: Getting Started
nav_order: 1
---
# Installation
{: .no_toc }
1. TOC
{:toc}
---
Note: To interoperate with OpenCV, OpenCV 3.x and above are preferred. OpenCV
2.x currently works but interoperability support may be deprecated in the
@ -11,22 +23,11 @@ Note: To make Mediapipe work with TensorFlow, please set Python 3.7 as the
default Python version and install the Python "six" library by running `pip3
install --user six`.
Choose your operating system:
Note: To build and run Android example apps, see these
[instructions](./building_examples.md#android). To build and run iOS example
apps, see these [instructions](./building_examples.md#ios).
- [Installing on Debian and Ubuntu](#installing-on-debian-and-ubuntu)
- [Installing on CentOS](#installing-on-centos)
- [Installing on macOS](#installing-on-macos)
- [Installing on Windows](#installing-on-windows)
- [Installing on Windows Subsystem for Linux (WSL)](#installing-on-windows-subsystem-for-linux-wsl)
- [Installing using Docker](#installing-using-docker)
To build and run Android example apps, see these
[instuctions](./building_examples.md#android).
To build and run iOS example apps, see these
[instuctions](./building_examples.md#ios).
### Installing on Debian and Ubuntu
## Installing on Debian and Ubuntu
1. Checkout MediaPipe repository.
@ -137,7 +138,7 @@ To build and run iOS example apps, see these
# Hello World!
```
### Installing on CentOS
## Installing on CentOS
1. Checkout MediaPipe repository.
@ -220,7 +221,7 @@ To build and run iOS example apps, see these
# Hello World!
```
### Installing on macOS
## Installing on macOS
1. Prework:
@ -352,7 +353,7 @@ To build and run iOS example apps, see these
# Hello World!
```
### Installing on Windows
## Installing on Windows
**Disclaimer**: Running MediaPipe on Windows is experimental.
@ -457,7 +458,7 @@ next section.
```
### Installing on Windows Subsystem for Linux (WSL)
## Installing on Windows Subsystem for Linux (WSL)
Note: The pre-built OpenCV packages don't support cameras in WSL. Unless you
[compile](https://funvision.blogspot.com/2019/12/opencv-web-camera-and-video-streams-in.html)
@ -580,7 +581,7 @@ cameras. Alternatively, you use a video file as input.
# Hello World!
```
### Installing using Docker
## Installing using Docker
This will use a Docker image that will isolate mediapipe's installation from the rest of the system.
@ -630,7 +631,7 @@ This will use a Docker image that will isolate mediapipe's installation from the
# Hello World!
```
4. Build Mediapipe [Android demos](./examples.md).
4. Build a MediaPipe Android example.
```bash
$ docker run -it --name mediapipe mediapipe:latest

View File

@ -1,12 +1,16 @@
# Troubleshooting
---
layout: default
title: Troubleshooting
parent: Getting Started
nav_order: 10
---
- [Native method not found](#native-method-not-found)
- [No registered calculator found](#no-registered-calculator-found)
- [Out Of Memory error](#out-of-memory-error)
- [Graph hangs](#graph-hangs)
- [Calculator is scheduled infrequently](#calculator-is-scheduled-infrequently)
- [Output timing is uneven](#output-timing-is-uneven)
- [CalculatorGraph lags behind inputs](#calculatorgraph-lags-behind-inputs)
# Troubleshooting
{: .no_toc }
1. TOC
{:toc}
---
## Native method not found
@ -55,7 +59,7 @@ running MediaPipe graph. This can occur for a number of reasons, such as:
For problem (1), it may be necessary to drop some old packets in older to
process the more recent packets. For some hints, see:
[How to process realtime input streams](how_to_questions.md).
[`How to process realtime input streams`].
For problem (2), it could be that one input stream is lacking packets for some
reason. A device or a calculator may be misconfigured or may produce packets
@ -63,7 +67,7 @@ only sporadically. This can cause downstream calculators to wait for many
packets that will never arrive, which in turn causes packets to accumulate on
some of their input streams. MediaPipe addresses this sort of problem using
"timestamp bounds". For some hints see:
[How to process realtime input streams](how_to_questions.md).
[`How to process realtime input streams`].
The MediaPipe setting [`CalculatorGraphConfig::max_queue_size`] limits the
number of packets enqueued on any input stream by throttling inputs to the
@ -122,14 +126,14 @@ each packet as early as possible. Normally the lowest possible latency is the
total time required by each calculator along a "critical path" of successive
calculators. The latency of the a MediaPipe graph could be worse than the ideal
due to delays introduced to display frames a even intervals as described in
[Output timing is uneven](troubleshooting.md?cl=252235797#output-timing-is-uneven).
[Output timing is uneven](#output-timing-is-uneven).
If some of the calculators in the graph cannot keep pace with the realtime input
streams, then latency will continue to increase, and it becomes necessary to
drop some input packets. The recommended technique is to use the MediaPipe
calculators designed specifically for this purpose such as
[`FlowLimiterCalculator`] as described in
[How to process realtime input streams](how_to_questions.md).
[`How to process realtime input streams`].
[`CalculatorGraphConfig`]: https://github.com/google/mediapipe/tree/master/mediapipe/framework/calculator.proto
[`CalculatorGraphConfig::max_queue_size`]: https://github.com/google/mediapipe/tree/master/mediapipe/framework/calculator.proto
@ -142,3 +146,4 @@ calculators designed specifically for this purpose such as
[`Timestamp::Done`]: https://github.com/google/mediapipe/tree/master/mediapipe/framework/timestamp.h
[`CalculatorBase::Close`]: https://github.com/google/mediapipe/tree/master/mediapipe/framework/calculator_base.h
[`FlowLimiterCalculator`]: https://github.com/google/mediapipe/tree/master/mediapipe/calculators/core/flow_limiter_calculator.cc
[`How to process realtime input streams`]: faq.md#how-to-process-realtime-input-streams

BIN
docs/images/accelerated.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 18 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 15 KiB

View File

Before

Width:  |  Height:  |  Size: 51 KiB

After

Width:  |  Height:  |  Size: 51 KiB

View File

Before

Width:  |  Height:  |  Size: 885 KiB

After

Width:  |  Height:  |  Size: 885 KiB

View File

Before

Width:  |  Height:  |  Size: 66 KiB

After

Width:  |  Height:  |  Size: 66 KiB

View File

Before

Width:  |  Height:  |  Size: 8.2 MiB

After

Width:  |  Height:  |  Size: 8.2 MiB

View File

Before

Width:  |  Height:  |  Size: 170 KiB

After

Width:  |  Height:  |  Size: 170 KiB

View File

Before

Width:  |  Height:  |  Size: 5.5 MiB

After

Width:  |  Height:  |  Size: 5.5 MiB

View File

Before

Width:  |  Height:  |  Size: 39 KiB

After

Width:  |  Height:  |  Size: 39 KiB

View File

Before

Width:  |  Height:  |  Size: 361 KiB

After

Width:  |  Height:  |  Size: 361 KiB

View File

Before

Width:  |  Height:  |  Size: 15 KiB

After

Width:  |  Height:  |  Size: 15 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.8 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 11 KiB

View File

Before

Width:  |  Height:  |  Size: 27 KiB

After

Width:  |  Height:  |  Size: 27 KiB

View File

Before

Width:  |  Height:  |  Size: 195 KiB

After

Width:  |  Height:  |  Size: 195 KiB

View File

Before

Width:  |  Height:  |  Size: 35 KiB

After

Width:  |  Height:  |  Size: 35 KiB

View File

Before

Width:  |  Height:  |  Size: 118 KiB

After

Width:  |  Height:  |  Size: 118 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 7.1 MiB

View File

Before

Width:  |  Height:  |  Size: 1.1 KiB

After

Width:  |  Height:  |  Size: 1.1 KiB

View File

Before

Width:  |  Height:  |  Size: 923 B

After

Width:  |  Height:  |  Size: 923 B

View File

Before

Width:  |  Height:  |  Size: 46 KiB

After

Width:  |  Height:  |  Size: 46 KiB

View File

Before

Width:  |  Height:  |  Size: 217 KiB

After

Width:  |  Height:  |  Size: 217 KiB

View File

Before

Width:  |  Height:  |  Size: 163 KiB

After

Width:  |  Height:  |  Size: 163 KiB

View File

Before

Width:  |  Height:  |  Size: 9.6 KiB

After

Width:  |  Height:  |  Size: 9.6 KiB

View File

Before

Width:  |  Height:  |  Size: 4.6 KiB

After

Width:  |  Height:  |  Size: 4.6 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.3 MiB

View File

Before

Width:  |  Height:  |  Size: 1.8 KiB

After

Width:  |  Height:  |  Size: 1.8 KiB

View File

Before

Width:  |  Height:  |  Size: 26 KiB

After

Width:  |  Height:  |  Size: 26 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 31 KiB

View File

Before

Width:  |  Height:  |  Size: 20 KiB

After

Width:  |  Height:  |  Size: 20 KiB

View File

Before

Width:  |  Height:  |  Size: 4.6 KiB

After

Width:  |  Height:  |  Size: 4.6 KiB

View File

Before

Width:  |  Height:  |  Size: 302 KiB

After

Width:  |  Height:  |  Size: 302 KiB

View File

Before

Width:  |  Height:  |  Size: 9.4 KiB

After

Width:  |  Height:  |  Size: 9.4 KiB

View File

Before

Width:  |  Height:  |  Size: 35 KiB

After

Width:  |  Height:  |  Size: 35 KiB

View File

Before

Width:  |  Height:  |  Size: 75 KiB

After

Width:  |  Height:  |  Size: 75 KiB

View File

Before

Width:  |  Height:  |  Size: 56 KiB

After

Width:  |  Height:  |  Size: 56 KiB

View File

Before

Width:  |  Height:  |  Size: 18 KiB

After

Width:  |  Height:  |  Size: 18 KiB

View File

Before

Width:  |  Height:  |  Size: 37 KiB

After

Width:  |  Height:  |  Size: 37 KiB

View File

Before

Width:  |  Height:  |  Size: 666 KiB

After

Width:  |  Height:  |  Size: 666 KiB

View File

Before

Width:  |  Height:  |  Size: 529 KiB

After

Width:  |  Height:  |  Size: 529 KiB

View File

Before

Width:  |  Height:  |  Size: 12 KiB

After

Width:  |  Height:  |  Size: 12 KiB

View File

Before

Width:  |  Height:  |  Size: 2.3 MiB

After

Width:  |  Height:  |  Size: 2.3 MiB

View File

Before

Width:  |  Height:  |  Size: 808 KiB

After

Width:  |  Height:  |  Size: 808 KiB

View File

Before

Width:  |  Height:  |  Size: 121 KiB

After

Width:  |  Height:  |  Size: 121 KiB

View File

Before

Width:  |  Height:  |  Size: 72 KiB

After

Width:  |  Height:  |  Size: 72 KiB

View File

Before

Width:  |  Height:  |  Size: 121 KiB

After

Width:  |  Height:  |  Size: 121 KiB

View File

Before

Width:  |  Height:  |  Size: 3.3 MiB

After

Width:  |  Height:  |  Size: 3.3 MiB

View File

Before

Width:  |  Height:  |  Size: 1.0 MiB

After

Width:  |  Height:  |  Size: 1.0 MiB

View File

Before

Width:  |  Height:  |  Size: 32 KiB

After

Width:  |  Height:  |  Size: 32 KiB

View File

Before

Width:  |  Height:  |  Size: 59 KiB

After

Width:  |  Height:  |  Size: 59 KiB

View File

Before

Width:  |  Height:  |  Size: 1.3 MiB

After

Width:  |  Height:  |  Size: 1.3 MiB

View File

Before

Width:  |  Height:  |  Size: 460 KiB

After

Width:  |  Height:  |  Size: 460 KiB

View File

Before

Width:  |  Height:  |  Size: 64 KiB

After

Width:  |  Height:  |  Size: 64 KiB

View File

Before

Width:  |  Height:  |  Size: 299 KiB

After

Width:  |  Height:  |  Size: 299 KiB

View File

Before

Width:  |  Height:  |  Size: 3.1 MiB

After

Width:  |  Height:  |  Size: 3.1 MiB

View File

Before

Width:  |  Height:  |  Size: 383 KiB

After

Width:  |  Height:  |  Size: 383 KiB

View File

Before

Width:  |  Height:  |  Size: 84 KiB

After

Width:  |  Height:  |  Size: 84 KiB

View File

Before

Width:  |  Height:  |  Size: 32 KiB

After

Width:  |  Height:  |  Size: 32 KiB

View File

Before

Width:  |  Height:  |  Size: 293 KiB

After

Width:  |  Height:  |  Size: 293 KiB

View File

Before

Width:  |  Height:  |  Size: 93 KiB

After

Width:  |  Height:  |  Size: 93 KiB

View File

Before

Width:  |  Height:  |  Size: 5.6 MiB

After

Width:  |  Height:  |  Size: 5.6 MiB

View File

Before

Width:  |  Height:  |  Size: 4.7 MiB

After

Width:  |  Height:  |  Size: 4.7 MiB

View File

Before

Width:  |  Height:  |  Size: 448 KiB

After

Width:  |  Height:  |  Size: 448 KiB

View File

Before

Width:  |  Height:  |  Size: 150 KiB

After

Width:  |  Height:  |  Size: 150 KiB

View File

Before

Width:  |  Height:  |  Size: 20 KiB

After

Width:  |  Height:  |  Size: 20 KiB

View File

Before

Width:  |  Height:  |  Size: 149 KiB

After

Width:  |  Height:  |  Size: 149 KiB

View File

Before

Width:  |  Height:  |  Size: 193 KiB

After

Width:  |  Height:  |  Size: 193 KiB

View File

Before

Width:  |  Height:  |  Size: 213 KiB

After

Width:  |  Height:  |  Size: 213 KiB

View File

Before

Width:  |  Height:  |  Size: 1.5 MiB

After

Width:  |  Height:  |  Size: 1.5 MiB

View File

Before

Width:  |  Height:  |  Size: 475 KiB

After

Width:  |  Height:  |  Size: 475 KiB

View File

Before

Width:  |  Height:  |  Size: 1.3 MiB

After

Width:  |  Height:  |  Size: 1.3 MiB

View File

Before

Width:  |  Height:  |  Size: 112 KiB

After

Width:  |  Height:  |  Size: 112 KiB

View File

Before

Width:  |  Height:  |  Size: 100 KiB

After

Width:  |  Height:  |  Size: 100 KiB

View File

Before

Width:  |  Height:  |  Size: 1004 KiB

After

Width:  |  Height:  |  Size: 1004 KiB

View File

Before

Width:  |  Height:  |  Size: 945 KiB

After

Width:  |  Height:  |  Size: 945 KiB

View File

Before

Width:  |  Height:  |  Size: 336 KiB

After

Width:  |  Height:  |  Size: 336 KiB

Some files were not shown because too many files have changed in this diff Show More