This is an Android app that uses machine learning to provide real-time face recognition. It leverages the Mobile FaceNet model, a lightweight neural network for face recognition that is optimized for mobile devices. The app is built with ML Kit and TensorFlow Lite, which provide powerful tools for image recognition and machine learning on mobile devices. The app's user interface is created using Jetpack Compose, a modern UI toolkit that streamlines the development of native Android apps.
The app offers the following features:
-
Real-time face detection and recognition: The app uses the device camera to detect and recognize faces in real time, allowing users to identify people quickly and easily.
-
Display of recognized person's name: When the app recognizes a person's face, it displays their name in real time, allowing users to confirm the person's identity without needing to ask them directly.
-
Option to add new faces to the recognition model: The app allows users to add new faces to the recognition model, so that the app can recognize more people over time.
-
Option to delete existing faces from the recognition model: The app also allows users to delete faces from the recognition model, so that they can maintain control over who the app can recognize.
-
Simple and intuitive UI: The app's user interface is designed with Jetpack Compose, a modern UI toolkit that reduces the amount of code needed to build native Android apps. The result is a simple, intuitive interface that is easy to use.
To use the app, you'll need the following:
- Android Studio Arctic Fox (2020.3.1) or newer
- Android SDK version 30 or newer
- Android Emulator or a physical Android device with a camera
To use the app, simply open it on your Android device or emulator, grant camera permissions, and point the camera at a person's face. If the person is recognized, their name will be displayed in real time. To add new faces to the recognition model, simply use the "Add face" button and follow the prompts. To delete existing faces from the recognition model, use the "Delete face" button and select the faces you want to delete.
The app uses the following technologies:
-
Mobile FaceNet model: A lightweight neural network for face recognition that is optimized for mobile devices. The model is trained to recognize faces with high accuracy, while using minimal computational resources.
-
ML Kit: Google's machine learning framework for mobile devices, used for face detection and recognition. ML Kit provides powerful tools for building machine learning models on mobile devices, making it easier to build apps that leverage machine learning.
-
TensorFlow Lite: An optimized version of the TensorFlow machine learning library, used for face recognition. TensorFlow Lite is designed to run efficiently on mobile devices, making it an ideal choice for building mobile machine learning apps.
-
Jetpack Compose: A modern UI toolkit for building native Android apps with less code. Jetpack Compose simplifies the process of building user interfaces, making it easier to build high-quality, responsive apps.
-
Kotlin: A modern programming language used for building Android apps. Kotlin is designed to be more expressive and concise than Java, making it easier to write readable, maintainable code.
This app was created by [KrishnaZyala]. It uses the following libraries and resources:
-
Mobile FaceNet: The Mobile FaceNet model used in the app was created by Sirius AI and is available on GitHub at https://github.com/sirius-ai/MobileFaceNet_TFLite_Android.
-
ML Kit: The app uses Google's ML Kit for face detection and recognition. More information about ML Kit can be found at https://developers.google.com/ml-kit.
-
TensorFlow Lite: The app uses TensorFlow Lite for face recognition. More information about TensorFlow Lite can be found at https://www.tensorflow.org/lite.
-
Jetpack Compose: The app's user interface was built using Jetpack Compose, a modern UI toolkit for building native Android apps. More information about Jetpack Compose can be found at https://developer.android.com/jetpack/compose.
This app is licensed under the MIT License. See the LICENSE file for more information.
Contributions to this project are welcome. If you find a bug or have a suggestion for a new feature, please create an issue on the project's GitHub page at https://github.com/KrishnaZyala/FaceRecognition. If you'd like to contribute code to the project, please create a pull request on the GitHub page.
If you have any questions or comments about the app, please feel free to contact.