Skip to main content
  1. Home
  2. Phones
  3. Mobile
  4. Web
  5. News

Microsoft’s Fetch! will classify your dog’s breed

Add as a preferred source on Google

Curious, but unable to find out what breed your dog is? There’s an app for that.

Microsoft’s creative lab is at it again with another iOS app showcasing its machine-learning technology, which identifies and distinguishes a dog’s breed, and it can even tell if you’ll be a good match.

Recommended Videos

Fetch! by Microsoft Garage, is a free app that uses your camera or photo library to classify a dog’s breed. So snap a picture of a dog, and the app will display the closest percentage of what your dog’s breed is — if your dog is a mix, simply tap the percentage and it will describe the top five breeds that could be the dog in the image. You can also hit the arrow on the bottom right to get a general description of the breed, what family the breed is suited for, as well as more pictures.

There’s a scrapbook feature as well, where you can access all your previous pictures.

Of course, the dog breed has to be “supported” by Microsoft’s database — which the company says has been curated by machine intelligence, and by hiring dog experts to verify images of breeds. The artificial intelligence in the app gets better the more you use it, and the Redmond company suggests users leave feedback if their dog breed was wrong, so the app can improve with every image it classifies.

If you take a picture of an object that’s not a dog, the app will say “No dogs found,” but will still offer a guess as to what it is. Pointing it at humans kicks in a “fun mode,” where the app will classify what type of dog you or your friends are, and also why. If it classifies you as a German Shepherd, as long as you use another photo, you won’t get the same result twice.

The app uses a technique called deep neural networks, which identifies the most subtle of differences in images to distinguish the right breed.

“It’s hard work to classify any number of breeds,” said Mitch Goldberg, development director at Microsoft Research, in the blog post announcement. “One of the things about using deep neural networks is that you don’t train it on a particular dog breed. You give it numerous images, and it, in the training process, determines what’s unique in each of the dogs.”

Thankfully, it’s not just restricted to iOS devices. Microsoft has set up a website that offers the same experience as the app, where you can upload photos to distinguish the breed type.

And what better place to try it, than at American Kennel Club’s Meet & Compete, or the Westminster Kennel Club Dog Show coming up in the next week. You can grab the app here.

Julian Chokkattu
Former Mobile and Wearables Editor
Julian is the mobile and wearables editor at Digital Trends, covering smartphones, fitness trackers, smartwatches, and more…
TSMC might set up a price hike that could come straight for your next phone, laptop, or tablet
Here's what TSMC's rumored 10% chip price increase actually means in dollar terms, and why your next phone or laptop could end up costing more.
TSMC Fab

My wallet flinched the second I saw the words "TSMC" and "price increase" in the same headline, and honestly, yours should too.

Turns out the company behind the silicon powering basically every flagship device out there, including Apple’s A-series and Qualcomm’s Snapdragon processors, is reportedly about to make all of it a little pricier.

Read more
Samsung’s free storage doubling for preorders is gone in Austria, and the replacement isn’t as sweet
Samsung's free storage doubling deal might just be history.
The back of the Galaxy Z Fold 7

If you've pre-ordered a Samsung foldable purely for the free storage bump, tomorrow's Unpacked event might sting a little.

For years, pre-ordering a new Galaxy meant automatically getting double the storage at no additional charge. Buy the 128GB model, walk away with 256GB for the same price. The same applied to the 256GB and 512GB models. However, that might change at the upcoming Galaxy Unpacked event. 

Read more
Apple fixes Hide My Email bug that exposed users’ real email addresses
Here's how Apple's Hide My Email flaw leaked real addresses for over a year, and why it only got fixed once the story went public.
apple-merging-sign-in-with-apple-hide-my-email-icloud+

Turns out the "Hide" part of Hide My Email wasn't doing its job quite as advertised, something that I covered early in July. Security researcher Tyler Murphy reported the flaw in June 2025, but despite Apple claiming it was resolved in March 2026, independent tests confirmed it remained exploitable, at least until July 3, 2026.

So how did this bug actually work?

Read more