The MLExplorer is an Machine Learning based app, in which user can recognize European Landmarks and can do Object Detection and Tracking.
- Implemented Firebase Authentication feature to develop Login and Registration Screen. So that only authenticated user can access the app.
- Implement Dagger Hilt Library to manage component Dependencies.
- Implement JetPack Compose Navigation Library to handle app Navigation, make app multi screen level.
- Built the UI using JetPack Compose toolkit adhered by Material 3 guidelines.
- Built upon MVVM architecture causing maintainability, reusability and testability.
- Kotlin to built backend logic.
- Android Studio FrameWork
- Implemented CameraX Library to access Device Camera component.
- Implement Kotlin Coroutines .
- For european landmark recognition I have integerated the tensorflow lite model which classifies or recognizes the european landmark.
- And if you want to know more here's the link - https://www.kaggle.com/models/google/landmarks/tfLite/classifier-europe-v1/1?tfhub-redirect=true
- For Object Detection and Tracking I have integerated the EfficientDet-Lite0 model. Why?
- Because it strikes a balance between latency and accuracy. It is both accurate and lightweight enough for many use cases.
- And for more information here's the link - https://developers.google.com/mediapipe/solutions/vision/object_detector
Authentication.mp4
Part2.mp4
- Now this app is not open for further contribution.
- And the Why its name is "The MLExplorer". Because the main motive behind developing this app is that I want to explore the ML field, I want to get an idea, an learning that how to integrate an Pre-Train model into an Android App. And what I can do with these models.
- So that's why I have named it The MLExplorer.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent