HighRes-Android

The task of recovering a high resolution (HR) image from its low resolution counterpart is commonly referred to as Single Image Super Resolution (SISR).

The model used here is ESRGAN (ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks). The TFLite model is converted from this implementation hosted on TF Hub. Demo images are from DIV2K dataset.

This sample automatically downloads TFLite JAR files and uses TFLite C API through Android NDK.

Requirements

• Android Studio 3.2 (installed on a Linux, Mac or Windows machine)
• An Android device, or an Android Emulator

Build and run

Step 1. Clone the TensorFlow examples source code

Clone the TensorFlow examples GitHub repository to your computer to get the demo application.

git clone https://github.com/tensorflow/examples

Step 2. Import the sample app to Android Studio

Open the TensorFlow source code in Android Studio. To do this, open Android Studio and select Import Projects (Gradle, Eclipse ADT, etc.), setting the folder to examples/lite/examples/super_resolution/android

Step 3. Download TFLite library

Open your terminal and go to the sample folder. Type './gradlew fetchTFLiteLibs' to run the download tasks. Use 'gradlew.bat' on Windows.

Step 4. Run the Android app

Connect the Android device to the computer and be sure to approve any ADB permission prompts that appear on your phone. Select Run -> Run app. Select the deployment target in the connected devices to the device on which the app will be installed. This will install the app on the device.

To test the app, open the app called TFL Super Resolution on your device. Re-installing the app may require you to uninstall the previous installations.

Future work:

• Use a distilled version to do video super resolution

Resources used:

• TensorFlow Lite
• ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks