Check the video to understand the code: https://youtu.be/dl7rCmvIyiI
-
git clone https://github.com/AarohiSingla/Object-Detection-Android-App.git
-
Train yolov8 model on custom dataset and export it in .tflite format. (Check train_export_yolov8_model.ipynb )
-
Open android_app folder.
-
Put your .tflite model and .txt label file inside the assets folder. You can find assets folder at this location: android_app\android_app\app\src\main\assets
-
Rename paths of your model and labels file in Constants.kt file. You can find Constants.kt at this location: android_app\android_app\app\src\main\java\com\surendramaran\yolov8tflite
-
Download and install Android Studio from the official website (https://developer.android.com/studio)
-
Once installed, open Android Studio from your applications menu.
-
When Android Studio opens, you'll see a welcome screen. Here, you'll find options to create a new project, open an existing project, or check out project from version control.Since you already have a project, click on "Open an existing Android Studio project".
-
Navigate to the directory where your project is located and select the project's root folder.
This project includes the andrroid app code from the following repository:
Special thanks to link-to-original-author-profile for their contribution.