[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-1628":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":25,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":43,"readmeContent":44,"aiSummary":45,"trendingCount":16,"starSnapshotCount":16,"syncStatus":46,"lastSyncTime":47,"discoverSource":48},1628,"mediapipe","google-ai-edge\u002Fmediapipe","google-ai-edge","Cross-platform, customizable ML solutions for live and streaming media.","https:\u002F\u002Fai.google.dev\u002Fedge\u002Fmediapipe",null,"C++",35579,6010,529,405,0,17,107,423,90,45,"Apache License 2.0",false,"master",true,[27,28,29,30,31,32,33,34,35,36,37,5,38,39,40,41,42],"android","audio-processing","c-plus-plus","calculator","computer-vision","deep-learning","framework","graph-based","graph-framework","inference","machine-learning","mobile-development","perception","pipeline-framework","stream-processing","video-processing","2026-06-12 02:00:30","---\nlayout: forward\ntarget: https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\ntitle: Home\nnav_order: 1\n---\n\n----\n\n**Attention:** *We have moved to\n[https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe)\nas the primary developer documentation site for MediaPipe as of April 3, 2023.*\n\n![MediaPipe](https:\u002F\u002Fdevelopers.google.com\u002Fstatic\u002Fmediapipe\u002Fimages\u002Fhome\u002Fhero_01_1920.png)\n\n**Attention**: MediaPipe Solutions Preview is an early release. [Learn\nmore](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fabout#notice).\n\n**On-device machine learning for everyone**\n\nDelight your customers with innovative machine learning features. MediaPipe\ncontains everything that you need to customize and deploy to mobile (Android,\niOS), web, desktop, edge devices, and IoT, effortlessly.\n\n*   [See demos](https:\u002F\u002Fgoo.gle\u002Fmediapipe-studio)\n*   [Learn more](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions)\n\n## Get started\n\nYou can get started with MediaPipe Solutions by by checking out any of the\ndeveloper guides for\n[vision](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fvision\u002Fobject_detector),\n[text](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Ftext\u002Ftext_classifier),\nand\n[audio](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Faudio\u002Faudio_classifier)\ntasks. If you need help setting up a development environment for use with\nMediaPipe Tasks, check out the setup guides for\n[Android](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fsetup_android), [web\napps](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fsetup_web), and\n[Python](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fsetup_python).\n\n## Solutions\n\nMediaPipe Solutions provides a suite of libraries and tools for you to quickly\napply artificial intelligence (AI) and machine learning (ML) techniques in your\napplications. You can plug these solutions into your applications immediately,\ncustomize them to your needs, and use them across multiple development\nplatforms. MediaPipe Solutions is part of the MediaPipe [open source\nproject](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmediapipe), so you can further customize the\nsolutions code to meet your application needs.\n\nThese libraries and resources provide the core functionality for each MediaPipe\nSolution:\n\n*   **MediaPipe Tasks**: Cross-platform APIs and libraries for deploying\n    solutions. [Learn\n    more](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Ftasks).\n*   **MediaPipe models**: Pre-trained, ready-to-run models for use with each\n    solution.\n\nThese tools let you customize and evaluate solutions:\n\n*   **MediaPipe Model Maker**: Customize models for solutions with your data.\n    [Learn more](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fmodel_maker).\n*   **MediaPipe Studio**: Visualize, evaluate, and benchmark solutions in your\n    browser. [Learn\n    more](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fstudio).\n\n### Legacy solutions\n\nWe have ended support for [these MediaPipe Legacy Solutions](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fguide#legacy)\nas of March 1, 2023. All other MediaPipe Legacy Solutions will be upgraded to\na new MediaPipe Solution. See the [Solutions guide](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fsolutions\u002Fguide#legacy)\nfor details. The [code repository](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmediapipe\u002Ftree\u002Fmaster\u002Fmediapipe)\nand prebuilt binaries for all MediaPipe Legacy Solutions will continue to be\nprovided on an as-is basis.\n\nFor more on the legacy solutions, see the [documentation](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmediapipe\u002Ftree\u002Fmaster\u002Fdocs\u002Fsolutions).\n\n## Framework\n\nTo start using MediaPipe Framework, [install MediaPipe\nFramework](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fframework\u002Fgetting_started\u002Finstall)\nand start building example applications in C++, Android, and iOS.\n\n[MediaPipe Framework](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fframework) is the\nlow-level component used to build efficient on-device machine learning\npipelines, similar to the premade MediaPipe Solutions.\n\nBefore using MediaPipe Framework, familiarize yourself with the following key\n[Framework\nconcepts](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fframework\u002Fframework_concepts\u002Foverview.md):\n\n*   [Packets](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fframework\u002Fframework_concepts\u002Fpackets.md)\n*   [Graphs](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fframework\u002Fframework_concepts\u002Fgraphs.md)\n*   [Calculators](https:\u002F\u002Fdevelopers.google.com\u002Fmediapipe\u002Fframework\u002Fframework_concepts\u002Fcalculators.md)\n\n## Community\n\n*   [Slack community](https:\u002F\u002Fmediapipe.page.link\u002Fjoinslack) for MediaPipe\n    users.\n*   [Discuss](https:\u002F\u002Fgroups.google.com\u002Fforum\u002F#!forum\u002Fmediapipe) - General\n    community discussion around MediaPipe.\n*   [Awesome MediaPipe](https:\u002F\u002Fmediapipe.page.link\u002Fawesome-mediapipe) - A\n    curated list of awesome MediaPipe related frameworks, libraries and\n    software.\n\n## Contributing\n\nWe welcome contributions. Please follow these\n[guidelines](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmediapipe\u002Fblob\u002Fmaster\u002FCONTRIBUTING.md).\n\nWe use GitHub issues for tracking requests and bugs. Please post questions to\nthe MediaPipe Stack Overflow with a `mediapipe` tag.\n\n## Resources\n\n### Publications\n\n*   [Bringing artworks to life with AR](https:\u002F\u002Fdevelopers.googleblog.com\u002F2021\u002F07\u002Fbringing-artworks-to-life-with-ar.html)\n    in Google Developers Blog\n*   [Prosthesis control via Mirru App using MediaPipe hand tracking](https:\u002F\u002Fdevelopers.googleblog.com\u002F2021\u002F05\u002Fcontrol-your-mirru-prosthesis-with-mediapipe-hand-tracking.html)\n    in Google Developers Blog\n*   [SignAll SDK: Sign language interface using MediaPipe is now available for\n    developers](https:\u002F\u002Fdevelopers.googleblog.com\u002F2021\u002F04\u002Fsignall-sdk-sign-language-interface-using-mediapipe-now-available.html)\n    in Google Developers Blog\n*   [MediaPipe Holistic - Simultaneous Face, Hand and Pose Prediction, on\n    Device](https:\u002F\u002Fai.googleblog.com\u002F2020\u002F12\u002Fmediapipe-holistic-simultaneous-face.html)\n    in Google AI Blog\n*   [Background Features in Google Meet, Powered by Web ML](https:\u002F\u002Fai.googleblog.com\u002F2020\u002F10\u002Fbackground-features-in-google-meet.html)\n    in Google AI Blog\n*   [MediaPipe 3D Face Transform](https:\u002F\u002Fdevelopers.googleblog.com\u002F2020\u002F09\u002Fmediapipe-3d-face-transform.html)\n    in Google Developers Blog\n*   [Instant Motion Tracking With MediaPipe](https:\u002F\u002Fdevelopers.googleblog.com\u002F2020\u002F08\u002Finstant-motion-tracking-with-mediapipe.html)\n    in Google Developers Blog\n*   [BlazePose - On-device Real-time Body Pose Tracking](https:\u002F\u002Fai.googleblog.com\u002F2020\u002F08\u002Fon-device-real-time-body-pose-tracking.html)\n    in Google AI Blog\n*   [MediaPipe Iris: Real-time Eye Tracking and Depth Estimation](https:\u002F\u002Fai.googleblog.com\u002F2020\u002F08\u002Fmediapipe-iris-real-time-iris-tracking.html)\n    in Google AI Blog\n*   [MediaPipe KNIFT: Template-based feature matching](https:\u002F\u002Fdevelopers.googleblog.com\u002F2020\u002F04\u002Fmediapipe-knift-template-based-feature-matching.html)\n    in Google Developers Blog\n*   [Alfred Camera: Smart camera features using MediaPipe](https:\u002F\u002Fdevelopers.googleblog.com\u002F2020\u002F03\u002Falfred-camera-smart-camera-features-using-mediapipe.html)\n    in Google Developers Blog\n*   [Real-Time 3D Object Detection on Mobile Devices with MediaPipe](https:\u002F\u002Fai.googleblog.com\u002F2020\u002F03\u002Freal-time-3d-object-detection-on-mobile.html)\n    in Google AI Blog\n*   [AutoFlip: An Open Source Framework for Intelligent Video Reframing](https:\u002F\u002Fai.googleblog.com\u002F2020\u002F02\u002Fautoflip-open-source-framework-for.html)\n    in Google AI Blog\n*   [MediaPipe on the Web](https:\u002F\u002Fdevelopers.googleblog.com\u002F2020\u002F01\u002Fmediapipe-on-web.html)\n    in Google Developers Blog\n*   [Object Detection and Tracking using MediaPipe](https:\u002F\u002Fdevelopers.googleblog.com\u002F2019\u002F12\u002Fobject-detection-and-tracking-using-mediapipe.html)\n    in Google Developers Blog\n*   [On-Device, Real-Time Hand Tracking with MediaPipe](https:\u002F\u002Fai.googleblog.com\u002F2019\u002F08\u002Fon-device-real-time-hand-tracking-with.html)\n    in Google AI Blog\n*   [MediaPipe: A Framework for Building Perception Pipelines](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.08172)\n\n### Videos\n\n*   [YouTube Channel](https:\u002F\u002Fwww.youtube.com\u002Fc\u002FMediaPipe)\n","MediaPipe 是一个跨平台的机器学习解决方案，专为实时和流媒体处理设计。它提供了强大的计算机视觉、音频处理及深度学习功能，基于C++开发，并支持图框架来构建可定制的ML管道。MediaPipe 支持多种设备上的部署，包括Android、iOS、Web、桌面以及边缘计算设备等，使得开发者能够轻松地将先进的机器学习特性集成到他们的应用中去。特别适合需要在移动或边缘设备上实现实时感知与理解的应用场景，如手势识别、面部追踪、语音分析等领域。",2,"2026-06-11 02:45:06","top_all"]