NutriVision is an Android app built with Kotlin and Jetpack Compose that allows users to capture or select a food image and instantly get its nutritional values (calories, protein, fat, carbs, etc.) using the Gemini Developer API.
The analyzed results are displayed and stored locally using Room Database for offline viewing.
Download and test latest version from release page👇
 |
 |
|---|---|
 |
 |
- Capture a food image from gallery
- AI-based food recognition using Gemini API
- Displays nutrition details (calories, protein, carbs, fat, fiber, minerals, vitamins, etc.)
- Saves analysis results locally (Room Database)
- View history of previous food analyses
- Clean modern UI with Jetpack Compose and Material 3
- Follows MVVM architecture pattern
- Language: Kotlin
- UI Framework: Jetpack Compose (Material Design 3)
- Architecture: MVVM + Repository pattern
- Database: Room Persistence Library
- Networking: Gemini Developer API (Google AI) using Firebase AI Logic SDK
- Serialization: kotlinx.serialization
- Asynchronous Tasks: Kotlin Coroutines + Flow
- User captures or uploads a food image.
- App sends the image + curated prompt to Gemini API for nutrition analysis.
- Gemini returns structured nutrition data in JSON format.
- App parses and displays the data on the Detail Screen.