How zarpec/visionartificial works
Project type and purpose
• This is a cross-platform mobile application built with React Native (TypeScript), targeting Android and iOS.
• Its purpose is to provide real-time computer vision assistance—object/obstacle detection, OCR, and audio guidance—for users with visual impairments.
File structure
• Top-level files & folders:
– App.tsx, index.js: entry points for the React Native JS application
– android/, ios/: native platform code and configuration
– __tests__/: Jest unit tests
– ml/ (under android/app/src/main/assets/models & android/app/src/ml): prebuilt ONNX models + TypeScript wrapper (detector.ts)
– export_model.py & yolov8n.onnx/pt: Python export script and model sources
– babel.config.js, metro.config.js, tsconfig.json, package.json: build and tooling configs
• Patterns:
– JS layer (React components & business logic)
– Native layer (Android: MainActivity.kt, Application.kt; iOS: AppDelegate.swift, Podfile)
– ML inference assets sitting alongside native resources
README for additional insights
• Core technologies: React Native, OpenCV, ONNX models, mobile device APIs (camera, mic, TTS)
• No external server: runs fully on-device
• Real-time audio orientation and OCR