Machine Learning on Android: Computer Vision

Heartbeat Collections

Computer Vision on Android

With recent advances in both Android device cameras and AI-accelerated chip technology, it’s no wonder that some of the most innovative, transformative use cases for machine learning on Android come in the form of computer vision.

From understanding scenes and creating artistic masterpieces to tracking human movement and even changing hair color, the promise of real-time computer vision on Android is real.

But it can be tough to know where to get started. That’s why we’ve worked to curate and assemble an authoritative collection of articles and tutorials that explore what’s possible with computer vision-based machine learning on Android.

Intro to Machine Learning on Android — How to convert a custom model to TensorFlow Lite

This post walks through how to convert a custom model to TensorFlow Lite in order to use it to run inference in Android.

— by Eric Hsiao

Intro to Machine Learning on Android (Part 2): Building an app to recognize handwritten digits with TensorFlow Lite

Create a model in TensorFlow trained on MNIST data, and then run inference on-device and in real-time on Android.

— by Eric Hsiao

TensorFlow Lite Model Maker: Build an Image Classifier for Android

Building machine learning models for edge devices just got a whole lot easier.

— by Anupam Chugh

Card Scanner on Android Using Firebase’s ML Kit and CameraX

Leveraging on-device text recognition to extract key information from business and other card types.

— by Lavanya Gaur

Image Recognition for Android with a Custom TensorFlow Lite Model

Deep learning on Android using Google Colab, transfer learning, and TFLite.

— by Ahmed Gad

Blink detection on Android using Firebase ML Kit’s Face Detection API

Learn how to use ML Kit’s on-device Face Detection API on Android to detect whether or not a user’s eyes are closed.

— by Harshit Dwivedi

Recognize Text in Images on Android with Firebase’s ML Kit

Identify text in images with the power of machine learning.

— by Danish Amjad

Training a TensorFlow Lite model for mobile using AutoML Vision Edge

Learn how to leverage Google’s AutoML Vision Edge training framework to train a custom image classification TensorFlow Lite model that’s ready for mobile deployment.

— By Harshit Dwivedi

CameraX: ‘The’ Machine Learning Camera Library for Android

Exploring the features of Android’s new CameraX library that make it the most convenient and versatile camera library for working with machine learning on Android.

— by Harshit Dwivedi

Real-Time Face Detection on Android with ML Kit

Using ML Kit, learn how build an app that can detect faces in an image in real-time.

— by Husayn Hakeem

Creating an Android app with Snapchat-style filters in 7 steps using Firebase’s ML Kit

Learn how to make your own Snapchat-style filters and integrate them into and Android app with Firebase’s ML Kit.

— by Zhang QiChuan

Using TensorFlow Lite and ML Kit to Build a “Pokédex” in Android

Learn how to build a custom TensorFlow Lite model with ML Kit that can predict Pokémon in real time.

— by Harshit Dwivedi

Using TensorFlow Lite and ML Kit to build custom machine learning models for Android

Learn how to build your own custom image classification model, convert it to TensorFlow Lite, and integrate it into an Android app.

— by Mohammed Rampurawala

Building a real-time object detection app on Android using Firebase ML Kit

At I/O 2019, the team behind Firebase ML Kit added two new APIs to their platform. This article is a simple implementation of the new real-time object detection API.

— by Harshit Dwivedi

Visual Recognition in Android Using IBM Watson

Learn what visual recognition means, how it works, and how to implement it into an Android app using IBM Watson.

— by Yash Soni

Image Classification on Android using a Keras Model Deployed in Flask

Learn how to perform image classification in an Android project by deploying a Keras model via a Flask server.

— by Ahmed Gad

Image Classification on Android using OpenCV

This tutorial uses the popular computer vision library OpenCV for building an image classifier that runs on Android devices.

— by Ahmed Gad