- Camera application to take jumping photos using mobile deep learning technique.

IT is An example Android application using TensorFLow Lite. 

This is a camera app that classifies images continuously using either a quantized Mobilenet model.

The app classifies frames in real-time, displaying the top most probable classifications. It also displays the time taken to detect the object.

As a result of classify, if the jump shot is more than 90%, it is automatically pictured and saved as an album automatically.

TensorFlow Lite -> Mobile Deep Learning Tool
TensorFlow Hub -> Deep Learning Models & Datasets Platform
Android Studio -> Mobile Application IDE
Camera2API -> Android camera API

-In the demo app, inference is done using the TensorFlow Lite Java API.
Mobile Classification: https://github.com/googlecodelabs/tensorflow-for-poets-2
Camera 2 API: https://github.com/googlearchive/android-Camera2Basic

To run the demo, a device running Android 5.0 ( API 21) or higher is required.
- Windows standard
  1. Run android Studio 
  2. Go to DOJumpShot/app/src/main/java/com/example/dojumpshot 
  3. Android OS Connect and run the device

Just press the ‘Picture’ button – application will keep checking the threshold value.

If jump accuracy exceeds the value of threshold, take a picture automatically. (Threshold: 0.9)

Taken photos will be saved in your gallery folder named ‘DOJumpShot’.