An intelligent Flutter application that automatically detects and processes images using ML Kit. Supports face detection with B&W filtering and document scanning with OCR text extraction.
 |
 |
 |
 |
 |
| Home Screen | Image Capture | Document Processing | OCR Text Extraction | Face Processing |
 |
 |
 |
 |
 |
| Face Result After | Face Result Before | Document Result | Detail View Document | Detail View Face |
You can watch the demo here: 👉 Watch Demo
- Smart Content Detection — Automatically identifies faces or documents using face + text analysis
- Face Processing — Applies black & white filter to all detected faces (supports multiple faces)
- Document Processing — Detects document region based on text and paper color, enhances contrast, and exports to PDF
- OCR Text Extraction — Recognizes text from documents — view, search inside text, copy to clipboard
- Processing History — Grid view of past results with thumbnails, metadata and detail screen
lib/
├── data/
│ ├── datasources/ # ML Kit & Image Processing
│ ├── models/ # Data Models
│ └── repositories/ # Repository Implementations
├── domain/
│ ├── entities/ # Business Entities
│ └── repositories/ # Repository Interfaces
└── presentation/
├── home/ # Home Screen
├── capture/ # Image Capture
├── processing/ # Processing Screen
├── result/ # Result Display
└── history_detail/ # Detail View
- Repository Pattern - Separation of data sources
- Dependency Injection - Using GetX bindings
- Reactive Programming - GetX state management
- SOLID Principles - Clean, maintainable code
-
State Management: GetX
-
Reactive state updates
-
Dependency injection
-
Simple navigation
-
No setState needed
-
Local Storage: Hive
-
Fast, lightweight NoSQL database
-
Type-safe data persistence
-
Perfect for metadata storage
-
Image Processing: Google ML Kit + image package
-
Face detection with bounding boxes
-
Text recognition (OCR)
-
Color-based edge detection
-
Custom processing algorithms
-
Face Detection
Uses two sensitivity levels (strict + relaxed fallback) to improve detection on small/distant/angled faces. -
Document Cropping
Expands from detected text region using paper color matching (instead of full contour detection) for speed and simplicity. -
Cropping Safety
Uses small padding (~10 px) + bounded expansion to avoid cutting important content.
| Category | Technology |
|---|---|
| Framework | Flutter 3.10.8 |
| Language | Dart 3.0+ |
| State Management | GetX 4.7.3 |
| ML & Computer Vision | Google ML Kit |
| Image Processing | Image Package 4.5.0 |
| PDF Generation | PDF 3.11.3 |
| Local Storage | Hive 2.2.3 |
| File Handling | Path Provider 2.1.2 |
dependencies:
get: ^4.7.3 # State management
google_mlkit_face_detection: ^0.13.2 # Face detection
google_mlkit_text_recognition: ^0.15.1 # OCR
image: ^4.5.0 # Image processing
image_picker: ^1.2.1 # Camera/Gallery
hive: ^2.2.3 # Local storage
hive_flutter: ^1.1.0 # Hive initialization
path_provider: ^2.1.2 # File paths
pdf: ^3.11.3 # PDF generation
open_file: ^3.5.11 # Open PDF viewer
intl: ^0.20.2 # Date formatting- Flutter SDK 3.10.8 or higher
- Dart SDK 3.0 or higher
- Clone the repository
git clone https://github.com/abir739/imageflow_flutter.git
cd imageflow_flutter- Install dependencies
flutter pub get- Run the app
flutter run- Tap the + button on home screen
- Choose Camera to take a photo or Gallery to select existing image
- App automatically detects content type (Face or Document)
- Watch real-time processing progress with status updates
For Face Processing:
- Toggle between Before/After to see B&W face filter
- Supports multiple faces in single image
For Document Scanning:
- View generated PDF
- Read extracted OCR text
- Search within text
- Copy text to clipboard
- Open PDF in external PDF viewer
- Browse all processed images in grid view
- Tap to view full details
- Long press to delete items