How mrniteshray/attendify works
Main Components
• Mobile Client (Android App): Represent the entire Android application with its key layers.
• UI Layer:
• Activities and Fragments (e.g., MainActivity, AdminActivity, FacultyFragment, SessionFragment, LoginActivity) provide the user interface.
• XML layouts (various activity and fragment layouts, navigation graphs, menus) for UI design.
• Core Logic & Helpers:
• Business logic for managing attendance, authentication, and session flows.
• Helper classes like FaceNetModel, TFLiteHelper that encapsulate face recognition and TensorFlow Lite interactions.
• API/Networking Layer:
• RetrofitClient and TwilioApi (used for sending WhatsApp notifications and interacting with external endpoints).
• Data Layer:
• Data models (AttendanceRecord, Classess, user) which structure the attendance and user data.
• External Services:
• Firebase services (Authentication, Firestore, Cloud Messaging) for backend storage and notifications.
• Google ML Kit for QR code scanning and geofencing functionalities.
• Twilio for sending alerts to parents and faculty.
Out the Relationships and Interactions
• Interaction flow starting with a Faculty member initiating an attendance session:
• The “Session Start” triggers the generation of a dynamic QR code and starts a face recognition check.
• The UI layer (activities/fragments) connects to the underlying helper modules where the triple-layered authentication occurs:
• QR Code scanning (using Google ML Kit) leads into geofencing (via Google Play Services).
• Face recognition components (using TFLite and the FaceNet model) verify the student's identity.
• Once the verification is successful, the attendance data is stored by interacting with Firebase services.
• In case of failed verification or for notifications, the business logic calls the API layer via Retrofit, which in turn integrates with the Twilio API to send WhatsApp notifications.
• Analytics and anti-spoofing measures should be shown as part of the business logic that monitors attendance patterns.
Architectural Patterns and Design Principles
• Separation of concerns: Clearly partition UI components, business logic, and data layers.
• Modular design: Each functional module (authentication, attendance, notifications) should be represented as independent blocks with defined interfaces.
• Real-time communication: Emphasize how Firebase Firestore and Cloud Messaging provide instant updates.
• Integration points: Make clear how external services (Firebase, Twilio, Google ML Kit) are incorporated.
Relevant Technologies and Frameworks
• Kotlin (for main programming logic)
• TensorFlow Lite with FaceNet (for face recognition)
• Google ML Kit (for QR code scanning and geofencing)
• Retrofit (for API communication)
• Firebase (for authentication, storage, messaging)
• Twilio API (for WhatsApp notifications)