Site navigation

Chinese AI Firm Develops Dog Facial Recognition App

Dominique Adams


dog nose close up

A startup backed by Alibaba says it has developed dog facial recognition software that uses nasal biometrics to identify individual dogs. 

Chinese artificial intelligence startup (AI), Megvii, claims to have created dog facial recognition software that can identify one dog from another using nasal biometric.

Many kennel clubs around the world are already using nose prints to keep track of dogs; however, this process involves dipping the dog’s nose in a liquid. Now with Megvii’s software, which uses a machine algorithm developed at the University of Utah, a dog’s snout can be scanned using only a smartphone camera.

To create the print requires a number of images taken of the nose from a variety of angles. The software then analyses the images and identifies key identification markers. The pet owner then uploads their images to a dedicated database to get identification certificates for their dogs.

Compared with previously recorded nose prints, Megvii’s nose prints are said to be 95% accurate. This method is pet-friendly and is cheaper than other methods of pet identification, such as microchipping and DNA sampling, according to Megvii.


The company boasts that it has reunited 15,000 pets with their owners via its app. But, the company says its technology can do more than just find lost dogs, it can also be used to monitor “uncivilised dog keeping”.

The software can be used as a tool by the government to catch and fine people who fail to control their dog properly. The move to crackdown on careless pet owners has been prompted by a rapid rise in complaints against pet owners.

If rolled out across China, which is home to an estimated 91 million pet cats and dogs, the app could help identify people who fail to clean up after their dogs or allow them to walk without leashes in public spaces.



Dominique Profile Picture

Dominique Adams

Marketing Content Manager, Trickle

Latest News

Artificial Intelligence Editor's Picks Machine Learning MedTech
Cybersecurity Editor's Picks
%d bloggers like this: