Snap. Analyze. Track. A modern Android application that uses Google's Gemini AI to recognize food from photos and track your daily nutrition intake instantly.

CaloriLens is a native Android application built demonstrating modern Android development practices. It leverages on-device CameraX for image capture and Cloud AI for image recognition to provide a seamless calorie tracking experience.

This project was built to showcase:

• Integration of Generative AI (multimodal) in mobile apps.
• Modern Android Architecture (MVVM, Unidirectional Data Flow).
• Clean UI/UX implementation with Jetpack Compose.

• AI-Powered Food Recognition: Instantly identifies food items and estimates serving sizes using Gemini 2.5 Flash.
• Real-time Analysis: Calculates macros (Calories, Protein, Carbs, Fat) on the fly.
• Smart Viewfinder: Custom camera interface with smart cropping to focus on the food.
• Daily Dashboard: Integrated calendar and progress rings to track daily goals.
• Local Persistence: Saves full meal history locally using Room Database.
• Premium UI: Dark mode support, smooth transitions, and glassmorphism elements.

• Language: Kotlin
• UI Toolkit: Jetpack Compose (Material 3)
• Architecture: MVVM (Model-View-ViewModel)
• AI Model: Google Gemini 2.5 Flash (via Generative AI SDK)
• Camera: CameraX
• Storage: Room Database & DataStore
• Networking: Retrofit & OkHttp
• Image Loading: Coil
• Concurrency: Coroutines & Flow

• Android Studio Koala or newer.
• A Gemini API Key from Google AI Studio.

1. Clone the repository:

   git clone https://github.com/soneryesilay/CaloriLens.git

2. Configure API Key:
   • Create a file named .env in the root directory.
   • Add your API key:

     GEMINI_API_KEY=your_actual_api_key_here

3. Build & Run:
   • Open the project in Android Studio.
   • Sync Gradle.
   • Run on an Emulator or Physical Device.

The app follows Clean Architecture principles:

• Data Layer: Repositories (FoodRepository, MealRepository) handle data fetching from AI and local DB.
• Domain/UI Layer: ViewModels (FoodViewModel, LogViewModel) manage state and business logic.
• UI: Composable functions rendering state (Success, Loading, Error).

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

Developed by Soner Yesilay