📱 ML Kit Pose Classification on iOS

This project brings pose classification to iOS using Google's ML Kit Pose Detection. It reimplements the pose classification logic (including the KNN algorithm) from the official Android sample project, filling the gap left by the lack of an iOS counterpart.

📂 Quickstart

This project is based on a fork of googlesamples/mlkit, focused on pose classification.

You can explore the relevant iOS implementation in this subfolder:
👉 ios/quickstarts/vision

🎯 Features

• Real-time human pose detection using ML Kit.
• K-Nearest Neighbors (KNN)-based pose classification.
• Custom pose embedding inspired by the Android example.
• Modular and extensible Swift implementation.

📍 Integration Overview

The PoseClassifierProcessor serves as the entry point for the classification module and is referenced inside CameraViewController.swift.

When the Detector.poseClassifier option is selected in the UI’s detector picker, the class is instantiated. Its detectPose(in image: MLImage...) method processes the results of the pose detection system and forwards them for classification.

📥 Installation

1. Clone the repo.
2. Open the Xcode project.
3. Install dependencies (e.g. via CocoaPods or Swift Package Manager):

   pod install