A New Way to Find a Lost Dog

FindingRover.com offers a new way to find your lost dog.

By Cari Jorgensen | Posted: August 8, 2014, 12:30 p.m. EDT

When I was in college I came home one day to discover that my Cocker Spaniel had gotten out. Apparently he thought he couldn’t leave the area in front of the door, which is where I found him. Luckily for me he hadn’t wandered and gotten lost.

But what if he had?

I would have spent hours, perhaps days, knocking on neighbors doors asking if they’d seen Rusty. I would have made flyers and stuck them all over the city. I probably would have hoped and prayed that a Good Samaritan would find him and bring him back.

Some of you may have done the same. Some may post photos and notices to your social media accounts. Others have had microchips implanted to search for your dogs if, God forbid, they ever get lost.

But now there’s a new, scientific way to search for lost dogs: facial recognition.


Dog Face 

John Polimeno founded FindingRover.com, a website that uses facial recognition technology to reunite lost dogs with their owners. Drs. John Schreiner and Steven Callahan of the Software Development Center at the University of Utah built the dog facial recognition technology that powers FindingRover.com.

Currently only servicing communities in San Diego County in southern California, the website contains a database of photographs from three shelters. When an owner of a lost pet uploads an image of their dog, the site’s technology attempts to match the image to theirs, looking for eight distinctive facial points. These markers include eye size and position of noses.

There has already been a successful match. Joanne Cox and her family lost their Shiba inu, Roxy. After she’d been missing for five days, they created a free account on FindingRover.com and uploaded a photo. That photo was soon matched with one from a shelter and within four hours of the match they had Roxy back.

After an owner uploads a photo of their lost dog, the program can work in one of two ways. Either that image is matched with the one a Good Samaritan who finds the dog uploads to the site, the owner receives a notice with the Good Samaritan’s information and a pickup is arranged; or dogs that come into the shelters are photographed and their images are matched with those in the database. In that scenario, the owner receives a phone call.

While accuracy is difficult to measure, a database of 100 dogs would have a top-three match 98 percent of the time. Polimeno is hoping to expand the website’s database to include additional shelters, veterinarians, dog groups and rescues.

FindingRover.com’s use of facial recognition is one of many ways to find lost dogs. How successful do you think it will be? Would you use it?

More Lost Dogs


0 of 0 Comments View All 0 Comments

Give us your opinion Give us your opinion on A New Way to Find a Lost Dog

Login to get points for commenting or write your comment below

First Name : Email :
International :
City : State :

Captcha Image

Get New Captcha

Top Products