Fritz AI Studio Now Supports Image Segmentation

Build, test, and deploy mobile-ready image segmentation models

We’re happy to report that Fritz AI Studio, our end-to-end development platform for mobile ML, now supports image segmentation use cases, in addition to object detection, pose estimation, and image labeling. This means that developers working with Fritz AI can build mobile-ready segmentation models using all the tools in the Studio — from generating an initial dataset all the way to monitoring and improving segmentation models in production.

Image segmentation is a computer vision technique used to understand what is in a given image at a pixel level. It’s different from image recognition, which assigns one or more labels to an entire image; and object detection, which localizes objects within an image by drawing a bounding box around them. Image segmentation provides more fine-grain information about the contents of an image.

For a more comprehensive look at image segmentation, check out our full guide

We can readily see the power of image segmentation when we consider its flexibility in photo editing software. From automatically separating backgrounds and foregrounds, to cutting out segmented people and objects, to creating responsive portrait modes, image segmentation offers a wide range of capabilities for these kinds of creativity tools. This is just one possibility, but it’s one that clearly indicates the potential of image segmentation.

Here’s a look at a demo from Chris, one of our engineers — a model that segments a face mask fast enough to work on live video, allowing users to change the color of their masks in real-time.

And one more from another of our engineers, Matt (they were having too much fun with this):

Getting Started

To start building with Fritz AI Studio and image segmentation, you can sign up for a Fritz AI account (if you haven’t already), which gives you access to all the tools you’ll need to build a cool portrait mode or your own green screen effect.

For current Fritz AI Studio users — to start building with image segmentation, you can log in as you normally would, and the appropriate configurations should be available to you.

If you need a little more help figuring out how to get started with the Studio, here are a couple additional resources:

  • Quickstart Guide: An accessible walkthrough for building, deploying, and testing a custom model.
  • Full tutorial: A detailed, step-by-step guide that takes you through the entire workflow
  • Video Demo: Instantly access a 17-min video walkthrough of Fritz AI Studio

We’re excited to see what you build — if you come up with something cool, leave us a note in the comments, tag @fritzlabs on Twitter, etc…you probably know how to find us by now 😏

Editor’s Note: Heartbeat is a contributor-driven online publication and community dedicated to exploring the emerging intersection of mobile app development and machine learning. We’re committed to supporting and inspiring developers and engineers from all walks of life.

Editorially independent, Heartbeat is sponsored and published by Fritz AI, the machine learning platform that helps developers teach devices to see, hear, sense, and think. We pay our contributors, and we don’t sell ads.

If you’d like to contribute, head on over to our call for contributors. You can also sign up to receive our weekly newsletters (Deep Learning Weekly and the Fritz AI Newsletter), join us on Slack, and follow Fritz AI on Twitter for all the latest in mobile machine learning.

Head of Community - Comet (comet.ml)