Introducing EyeEm Vision

By EyeEm Team - 3 min read

Today’s most advanced image recognition technology can detect the concepts and emotions in photos to get you discovered.

For the last two years, we’ve studied, researched, and developed technology to answer one of the most pressing problems for both photographers and buyers: getting their photos discovered and discovering photos. From this idea, EyeEm Vision was born, a computer vision engine that tags photos based on what it detects in the image.

We saw that the biggest challenge for photographers wasn’t tagging based off of the objects in their photos, but rather the ideas behind them. EyeEm Vision takes a stride beyond autotagging – it has the power to recognize a photographer’s intention, whether that be a concept or an emotion. The result? EyeEm Vision surfaces millions of photos that may have never been discovered, instantly connecting photographers and photo buyers through the ideas that they love to share.

EyeEm Vision is now available in beta. Be the first to try it.

Cropped image of hand holding compass against sky

Your Photos Seen Now More Than Ever

We made EyeEm Vision to grow the EyeEm community, give photographers exposure, and deliver the best content for Market buyers, all while furthering our quest to better understanding photos. With its success, your photos will now be seen faster and more effectively than ever before. It has the power to suggest tags that you might not have known buyers and photographers were searching for, and by taking the guesswork out for you, makes your photos infinitely more discoverable from the moment you upload.

Photo by @composethestory

The Magic Behind the Technology

Using a mixture of deep learning and the knowledge gathered from millions of images shared on the EyeEm community, EyeEm Vision reads images in a highly advanced way. Beyond detecting objects in a photo, it can actually ‘look’ at the context of an image and interpret the ideas and emotions behind it. For example, EyeEm Vision will suggest tags like ‘mountain,’ ‘sky,’ or hiker,’ but also for abstract concepts like ‘Freedom,’ ‘From the top,’ or ‘Looking forward.’

This is all thanks to a sophisticated algorithm that our scientists diligently trained with data. They essentially taught EyeEm Vision how to read an image and what concepts are important to both photographers and photo buyers alike. And the learning doesn’t end there, we’re now looking into the usage of the best 360 camera. As our community of 15 million photographers begin to use EyeEm Vision, it will continuously learn, growing ever more accurate with every photo uploaded.

We’re bridging the gap between technology and real photography, making data more human and your photos smarter.

Header image by @jimbobombo