StickerSense 🏷️

StickerSense is a local-first, smart sticker organizer for mobile devices. It leverages on-device Machine Learning (ML) to automatically tag and index your sticker collection, making it instantly searchable without compromising your privacy.

---

✨ Features

• Dual AI Modes 🤖:
  • Cloud API (default): Uses Gemma 3 27B via Google AI API for maximum quality (30 RPM limit)
  • On-Device: Downloads Gemma 3 Nano E4B (~4.4GB) for unlimited, offline AI tagging
• Hybrid AI Indexing: Combines AI analysis with on-device OCR for text extraction from memes.
• WhatsApp Import 📥 (Android): Folder picker to import your existing WhatsApp sticker collection with automatic AI tagging.
• Smart Fallback: Automatically reverts to local ML Kit (Labeling, Face Detection, Translation) if AI fails or is offline.
• Full-Text Search 🔍: Instant search across all indexed stickers using SQLite FTS5.
• Custom Keyboard (Coming Soon): Access your stickers directly from any chatting app via a custom system keyboard.
• Modern UI: Clean, responsive layout built with Flutter and Riverpod.nsures 100% offline functionality.
• Fast Search: Built on SQLite (Drift) with FTS5 for lightning-fast full-text search capabilities.
• Custom Keyboard (Coming Soon): Access your stickers directly from any chatting app via a custom system keyboard.
• Modern UI: Clean, responsive layout built with Flutter and Riverpod.

🔑 Setup (Gemma AI)

1. Get your API Key from Google AI Studio.
2. Create a .env file in the root directory.
3. Add your key: GEMINI_API_KEY=your_api_key_here.

Note: The free tier has usage limits. The app automatically handles rate limiting when "Più preciso" mode is enabled. For current limits, see Google AI API Pricing.

🛠️ Tech Stack

• Framework: Flutter
• State Management: Riverpod
• Database: Drift (SQLite)
• ML & AI:
  • Google Gemma 3 27B (Cloud-based semantic analysis)
  • Google ML Kit (On-device: Image Labeling, OCR, Face Detection, Translation)
• Architecture: Clean Architecture + Feature-First structure.

🚀 Getting Started

Prerequisites

• Flutter SDK (3.0 or later)
• Dart SDK
• Android Studio / Xcode

Installation

1. Clone the repository

   git clone https://github.com/yourusername/stickersense.git
   cd stickersense

2. Install dependencies

   flutter pub get

3. Run Code Generation This project uses build_runner for Riverpod and Drift code generation.

   dart run build_runner build -d

4. Run the App

   flutter run

📂 Project Structure

lib/
├── core/                   # Core functionality (Database, Services)
│   ├── database/           # Drift database schema and connection
│   └── services/           # ML services (Image Indexer)
├── features/               # Feature-based organization
│   └── sticker/            # Sticker management feature
│       ├── data/           # Repositories and Data Sources
│       └── presentation/   # UI widgets and Controllers
└── main.dart               # Application entry point

🤝 Contributing

Contributions are welcome! Please check the CONTRIBUTING.md file for guidelines on how to proceed.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.