https://github.com/tito/experiment-tensorflow-lite

https://github.com/tito/experiment-tensorflow-lite

Last synced: over 1 year ago
JSON representation

• Host: GitHub
• URL: https://github.com/tito/experiment-tensorflow-lite
• Owner: tito
• Created: 2020-10-04T18:16:24.000Z (almost 6 years ago)
• Default Branch: main
• Last Pushed: 2020-10-04T18:19:41.000Z (almost 6 years ago)
• Last Synced: 2025-04-18T00:55:24.913Z (over 1 year ago)
• Language: Python
• Size: 786 KB
• Stars: 26
• Watchers: 2
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# TensorFlow lite with Kivy

This is an experimental project, no support will be done.

Works only on Android for now.

![screenshot](https://github.com/tito/experiment-tensorflow-lite/blob/main/screenshot.jpg?raw=true)

It uses:

- Kivy android camera to grab RGBA pixels

- Pillow to convert the image to the tensor input format

- TensorFlow lite Android libraries to detect

This example is mostly based on the Image classification from TensorFlow

lite example repository.

## Create a model

- Train your model https://teachablemachine.withgoogle.com/

- Export as a Quantized model for TensorFlow lite

- Copy it into the app, ajust the paths.