A little-known German start-up may have just made it a lot easier to search for photos online.
EyeEm, a photo-sharing service started in 2011 that has drawn parallels to Instagram, announced new technology in Brooklyn on Friday that uses a sophisticated algorithm and machine learning to analyze the details of online photos.
The technology, called EyeVision, automatically scans images and tags them with certain keywords, from “landscape” and “New York” to the perceived emotions of people in each photo, which makes them easier to find through web searches.
While other companies have tried similar techniques to categorize online images, the German start-up’s efforts — which comes after roughly three years of development — are based on analyzing millions of photos already shared on EyeEm’s photo-sharing social network. The company has roughly 15 million users compared to about 300 million on Instagram.
Combined with the company’s machine-learning techniques, EyeEm’s search algorithm adapts over time to better understand what is part of each uploaded image, making it easier to find specific photos online, according to the company’s chief executive.
“We’re teaching the algorithm to identify what is a beautiful photo,” said Florian Meissner, the chief executive of EyeEm, which has raised money from influential West Coast backers, including Valar Ventures, a company backed by Peter Thiel. “We’ve proved that our technology works. It could be applied to almost all photos online.”
EyeEm’s foray into photo search comes as many of the world’s largest tech companies, like Google, Facebook and Amazon, are increasingly looking for ways to include images and other non-text responses in people’s online search queries.
Not everything has gone to plan. Google found itself in hot water earlier this year when its own photo-tagging service mistakenly labeled photos of black people as “gorillas.”
While Mr. Meissner said EyeEm was discussing potential partnerships to license its photo-search technology to others, some of the start-up’s investors said making it easier to find photos online could be applied to a range of online industries.
“We’re talking about the ability search and group images automatically,” said Stefan Glaenzer, a partner at London-based Passion Capital, which was one of EyeEm’s first backers. “For many companies, there needs to be a way to structure that photo data in a user-friendly way.”
The technology is by no means perfect, though. When a photo of the iconic Sydney Opera House was uploaded into EyeEm’s system, for example, the algorithm automatically tagged the image with many useful searchable terms, but failed to include either “Sydney” or “Opera House.”
EyeEm’s executives admit the technology needs a few tweaks, but say it is sophisticated enough to analyze the type of emotions — and even photographic style — present in each uploaded image.
“People need a way to structure their photos, they need it to be simple,” said Lorenz Aschoff, a co-founder of EyeEm. “Our vision is to automate that completely.”
For now, the start-up plans to use its photo-searching technology on its own online image marketplace, which allows people to upload photos and sell them to companies looking for stock imagery. EyeEm also has a partnership with Getty Images so that people’s smartphone photos are included in the company’s existing collection of stock images.
The online market, which opened earlier this year and is available in the United States and certain European countries, already has 60 million uploaded photos, which will now be scanned using EyeEm’s search algorithm. The company also plans to include geolocation data, if people give their consent, as part of the search results.
“My photos are super mundane, but it’s fun when I see someone has bought them through EyeEm,” said Danielle Reid, a Berlin-based product designer who has earned almost $700 through the company’s online marketplace. “I’m not a professional photographer, but it’s always nice to make a little extra money.”
By Mark Scott, posted on bits.blogs.nytimes.com