Skip to Content

A Photo Service That Understands the Contents of Your Images

Everpix organizes photos after analyzing them with software that can detect things such as animals, outdoor scenes, and people.
March 8, 2013

Browsing digital photos usually means scrolling through them chronologically, unless they have been sorted into folders and collections. This week a startup company called Everpix began offering an alternative: a system that uses machine vision software to analyze each photo for its content so that photos can be browsed using categories such as “city,” “animals,” “people,” and “nature.”

hands holding tablet
Image recognition: Everpix’s iPad app automatically categorizes images after identifying their contents.

The category-based view, called Explore, is now a feature of the company’s iPad and iPhone apps. It joins an existing feature of those apps and the company’s website that provides a way to browse the “highlights” from a collection of photos in a particular year. Those highlights are compiled into a scrollable collage by software that looks for signals suggesting that a photo is high-quality and interesting.

Thanks in part to the ubiquity of smartphone cameras, many people’s digital photo collections now contain thousands of images. At that size, they are becoming unmanageable with conventional tools such as Apple’s iPhoto, says Pierre-Oliver Latour, CEO and cofounder of Everpix, which is based in San Francisco. “We’re building something to solve this big problem that is coming where people are going to have too many photos and they begin to miss out on them and neglect them,” he says.

Many people have already reached that point, says Latour. Since launching quietly in 2011, Everpix has attracted tens of thousands of users to its service, which until this week cost at least $49 a year. The average new user uploads more than 10,000 photos, from sources including Windows and Apple PCs, mobile devices, and Facebook accounts, says Latour. A new free tier of the service, launched this week, offers a user access to just the last 12 months’ worth of photos; paying $49 a year allows access to an unlimited number.

Most photo organizing software relies on time stamps and user-created categories and folders, although some, such as Apple’s iPhoto, Google’s Picasa, and Facebook, use facial recognition as a way to find photos of particular people.

Everpix does not use facial recognition, but in a demonstration at the company’s offices, Latour and cofounders Kevin Quennesson and Wayne Fan showed evidence that their software understands much more than the categories its software now exposes to users. The software can identify when an uploaded image contains plants, babies, animals, water, or snow, for example. A database of word meanings has been integrated into the system so it can understand other ways to refer to the label it’s applied to a photo.

The image analysis software was trained by having many thousands of images labelled by crowdsourced workers, and the new Explore feature correctly categorises photos most of the time. When it doesn’t, a user can provide feedback to help Everpix train its software further.

Latour says future features will take advantage of the deeper understanding his company’s technology can mine from photos. During the demonstration, a search interface developed for internal research purposes was able to accurately find photos in response to queries such as “city photos with crowds from April 2012” and “city photos with people that are close to the camera.” Latour wouldn’t say whether that same interface would later appear in the Everpix website or apps.

Keep Reading

Most Popular

It’s time to retire the term “user”

The proliferation of AI means we need a new word.

Sam Altman says helpful agents are poised to become AI’s killer function

Open AI’s CEO says we won’t need new hardware or lots more training data to get there.

An AI startup made a hyperrealistic deepfake of me that’s so good it’s scary

Synthesia's new technology is impressive but raises big questions about a world where we increasingly can’t tell what’s real.

Taking AI to the next level in manufacturing

Reducing data, talent, and organizational barriers to achieve scale.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.