kukumaluCN / MPIrisTracker

This is an ios framework for Iris detection based on MediaPipe

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

MPIrisTracker

  • An iOS Framework that enables developers to use iris detection & capturing face mesh based on MediaPipe.

Key Features

  • enable to use MediaPipe Iris Detection in Swift App
  • Acquire iris tracking from front camera image
    • face mesh
    • iris landmark

Example Projects

To try the example project, simply clone this repository and open the examples folder

Develop Environment

  • Language: Swift
  • Xcode: Xcode version 12.5
  • Using Libralies:

Get Started

Installation

  1. Add MPIrisTracker.framwork in ./output folder to your App Xcode Project on Xcode.
  2. In Xcode, select File > New > File...
  3. Create temporary Objective-C File (this file will be deleted after creating bridging header file).
  4. (Xcode automatically creates xxx-Bridging-Header.h)
  5. Append #import <MPIrisTracker/MPIrisTracker.h> into xxx-Bridging-Header.h
  6. (if needed) Select Embed & Sign at MPIrisTracker.framework
    スクリーンショット 2021-06-01 22 01 44
  7. (if needed) Select No at Enable Bitcode
    スクリーンショット 2021-06-01 22 04 27

Permission Settings

  1. Make sure you add the usage description of the camera in the app's Info.plist.
<key>NSCameraUsageDescription</key>
<string>MPIrisTracker</string>

How to compile framework

Install MediaPipe

  1. Follow official MediaPipe documents

Compile

  1. Copy files in ./source folder to MediaPipe installation path (e.g. MEDIAPIPE_PATH/mediapipe/MPIrisTracker/).
  2. cd MEDIAPIPE_PATH
  3. build -c opt --config=ios_arm64 mediapipe/MPIrisTracker:MPIrisTracker

Reference

Licence

MIT

Author

Yuki Yamato [ukitomato]

About

This is an ios framework for Iris detection based on MediaPipe

License:MIT License


Languages

Language:Swift 43.8%Language:Objective-C++ 36.4%Language:Objective-C 14.0%Language:Starlark 5.8%