Image Classification, Image Captioning and LLM inference with LiteRT
Image Classification & Image Captioning with LiteRT
This is a fully functional Android application built with Kotlin and Jetpack Compose that showcases how to integrate LiteRT for on-device image classification and image captioning. It also demonstrates how to use MediaPipe for on-device large language model (LLM) integration, enabling natural language processing directly on the device.
You can read more about this repository in the blog post [Exploring On-Device AI for Android: Image Classification, Captioning & LLM Integration][blog-post]
The project follows modern Android development best practices and serves as a practical reference for developers interested in integrating LiteRT and MediaPipe into their own apps.
Features
Libraries
|
|
[litert]: https://github.com/google-ai-edge/LiteRT
[genai]: https://ai.google.dev/edge/mediapipe/solutions/genai/llm_inference/android
[core]: https://developer.android.com/jetpack/androidx/releases/core
[mvvm]: https://developer.android.com/topic/libraries/architecture/viewmodel
[hilt]: https://developer.android.com/training/dependency-injection/hilt-android
[compose]: https://developer.android.com/jetpack/compose
[retrofit]: http://square.github.io/retrofit
[moshi]: https://github.com/square/moshi
[coil]: https://coil-kt.github.io/coil/compose
[blog-post]: https://medium.com/proandroiddev/exploring-on-device-ai-for-android-image-classification-captioning-llm-integration-23080f965e65
