Category: Articles

Sensor fusion is one of the key aspects of self-driving cars.

If we take a look at the 4 main elements of self-driving vehicles, we can categorize sensor fusion a part of both the perception and the localization world. The reason is simple: both these disciplines use multiple sensors to function.

For example, perception is using cameras, LiDARs, and RADARs to detect obstacles’ classes, positions, and velocities with high accuracy. Localization, on the other hand, fuses GPS, LiDAR, and camera data to get an accurate position with centimeter-level accuracy.

Continue reading LiDAR and Camera Fusion in Self-Driving Cars

Today we’re excited to announce the Fritz AI Dataset Collection System, an important piece in our effort to help ML teams and developers generate, collect, and use data for mobile machine learning projects.

Data is a critical component to AI workflows, and the best data comes from the real world, where your models are (or will be) deployed.

For mobile machine learning projects, that means a user’s phone. The new Dataset Collection system makes it easy to gather this data through the same convenient SDK that delivers and runs your models.

Continue reading “Introducing the Fritz AI Dataset Collection System”

When the dust settled on a recent mobile machine learning project, we had accumulated 392 different model checkpoints. With numerous architectures to test, dozens of hyperparameters to sweep, and multiple on-device formats to support, models piled up quickly.

Staying organized and creating efficient workflows were the keys to success. We knew we needed to make this process easier—so we developed a set of command line tools to help.

Continue reading “Organizing mobile machine learning projects with the Fritz CLI”

Apple gave its Vision framework a major boost during WWDC 2019 by adding a lot of advancements. From expanding the number of classes (the term taxonomy is used for this) of its image classification requests to improvements in its face technology and text recognition requests, Apple is bringing in some really interesting improvements in computer vision for iOS.

Continue reading Vision Image Similarity Using Feature Prints in iOS

Though it didn’t get a ton of stage time during the keynote at this year’s WWDC, there’s a lot to be excited about in the latest iteration of Apple’s machine learning framework, Core ML 3.

In this post, I’ll highlight the biggest changes to the software and discuss their implications for developers and machine learning engineers.

Continue reading “What’s new in Core ML 3”

Dominos is running a campaign to get you to…well, think about Dominos some more. Take a picture of a pizza—any pizza—and get free points towards a free Dominos pie.

How can they do this? Does a person look at every photo taken and determine if it’s pizza or not? Fortunately, they do not:

Continue reading Recreate Dominos Points for Pies app on iOS with Fritz Image Labeling

Developers using Fritz AI can now add tags and metadata to on-device machine learning models. Models can be queried by tags and loaded dynamically via the iOS and Android SDKs, giving you more control over distribution and usage.

Deliver models to users based on hardware, location, software environment, or any other attribute.

Continue reading “Distributing on-device machine learning models with tags and metadata”

Neural networks have many practical use cases, from identifying street signs and people in self-driving cars to recognizing text and transcribing audio.

But one of the more fun use cases is applying the “style” of one image to another. Neural networks can be trained to tease apart the content of a painting from the style of a painting.

The results can then be applied to any image. The result is a set of fun photo filters that replicate famous paintings.

Continue reading “Real-Time Style Transfer for iOS— Transform your photos and videos into masterpieces”

There are now a bunch of off-the-shelf tools for training artistic style transfer models and thousands of open source implementations. Most use a variation of the network architecture described by Johnson et al to perform fast, feed-forward stylization.

As a result, the majority of the style transfer models you find are the same size: 7MB. That’s not an unreasonably large asset to add to your application, but it’s also not insignificant.

Research suggests that neural networks are often way larger than they need to be—that many of the millions of weights they contain are insignificant and needlessly precise. So I wondered: What’s the smallest model I can create that still reliably performs style transfer?

Continue reading “Creating a 17 KB style transfer model with layer pruning and quantization”