Detect obstacles in front of the user and warn them with voice support. Coarse classification categories supported: home goods, fashion goods, food, plants, and places.
Recognize text in front of the user and read it out loud.
- Anak Agung Gde Pradnyana
- Ari Ziddan Nugraha
- Muhammad Luthfi Khusyasy
- Raditya Aydin
- Flutter allows us to do fast prototyping and development.
- Save a lot of time and cost by using built-in widgets and packages available in the Flutter community.
- We use Firebase because it is easy to integrate with Flutter.
- Firebase also provides a lot of features and allows for easy scaling.
- We used Firebase Authentication to authenticate users. And planning to use Firebase ML to deploy and manage our machine learning model.
- TensorFlow Lite to run machine learning models on mobile devices for real-time object detection.
- Google ML Kit to use pre-trained machine learning models from Google for object detection and text recognition.
- We are planning to build our own model for the specific use case of detecting obstacles.
If you just want to try this app without running the development environtment, kindly download the apk using this download link
- Flutter 3.16.7
- Android SDK
- Android Emulator
- Clone the repository
- Run
flutter pub get
on terminal - Launch your emulator or connect your device
- Run debug using
F5
orflutter run