A privacy-first iOS app that generates skeleton pose overlays from photos using on-device human pose detection.
PoseDetect is a mobile application that allows users to capture or select images and generates a skeletal outline visualization of detected human poses. All processing happens locally on-device, ensuring complete privacy with zero data collection or internet connectivity required.
The app leverages Apple's native frameworks to perform real-time pose detection, making it ideal for fitness tracking, motion analysis, or creative applications where privacy is paramount.
- On-Device Processing - All pose detection runs locally with no cloud dependencies
- Privacy-First Design - Zero data collection, no internet required
- Skeleton Visualization - Generates clear skeletal overlays on captured images
- SwiftData Integration - Persistent local storage for detected poses
- Universal Support - Works on both iPhone and iPad
| Category | Technology |
|---|---|
| Language | Swift 5.0 |
| UI Framework | SwiftUI |
| Data Persistence | SwiftData |
| Platform | iOS 18.2+ |
| Build System | Xcode 16.2+ |
- Xcode 16.2 or later
- macOS with Xcode command line tools
- iOS device or simulator running iOS 18.2+
# Clone the repository
git clone https://github.com/coleschaffer/PoseDetect.git
# Open in Xcode
open PoseDetect/HumanPoseDetection.xcodeproj- Select your target device (simulator or physical device)
- Press
Cmd + Rto build and run - Grant camera permissions when prompted
PoseDetect/
├── HumanPoseDetection/
│ ├── HumanPoseDetectionApp.swift # App entry point with ModelContainer
│ ├── ContentView.swift # Main UI with navigation
│ ├── Item.swift # SwiftData model
│ ├── Assets.xcassets/ # App icons and colors
│ └── Preview Content/ # SwiftUI preview assets
├── HumanPoseDetectionTests/ # Unit tests
└── HumanPoseDetectionUITests/ # UI tests
This app is designed with privacy as a core principle:
- No data leaves your device
- No analytics or tracking
- No account required
- Works completely offline
For support, contact: [email protected]
MIT License