ML Kit for mobile apps

ML Kit is a mobile SDK that brings Google’s machine learning expertise to Android and iOS apps in a powerful yet easy-to-use package.

You need not be an expert in machine learning to write applications that use machine learning concepts. With ML Kit and a few lines of code, you can implement the functionality you need without learning anything about neural networks or model optimizations. However, experienced machine learning developers can also use the ML Kit SDK for integrating custom TensorFlow Lite models.

Although ML Kit is still in it’s early phases, common features are quite advanced and production ready. You can recognize text, detect faces, identify landmarks, scan barcodes and label images. You simply need to feed your data to the API and receive the information you need.

And all the computation is done can be done on your mobile device, although cloud option is also available. ML Kit combines Google’s ML technologies such as Cloud Vision API, TensorFlow Lite and Android Neural Networks API in a single SDK with the option of processing on the cloud or with mobile-optimized on-device models when you are offline.

All this, is for free till now for all on-device computation. There might be separate pricing for on-cloud processing. As more and more features are added to the ML Kit SDK, the base APIs can also cover more use cases in vision, speech and text fields. Also, the SDK and the models will gradually become more and more reliable.

ML Kit currently has great support for the following functionalities:

  1. Text Recognition for recognizing and extracting text from images.
  2. Face Detection for detecting faces and facial landmarks.
  3. Barcode Scanning for scanning and processing barcodes.
  4. Image Labelling for identifying objects, locations, activities, animal species, products and more.
  5. Landmark Recognition for identifying popular landmarks in images.

All you need is to set up a Firebase project and a few lines of code.

Leave a Reply

Your email address will not be published. Required fields are marked *