If internet memes have proven anything, it is that there is an oddly specific stock photo out there for almost any situation, but finding the right photo out of millions can take a lot of time and effort. Stock photo company Shutterstock Inc. wants to make that process easier with its new Composition Aware Search, a search tool powered by machine learning that allows users to search photos by their composition.

According to Shutterstock, the artificial intelligence that powers its new search feature is capable of understanding not only what is contained in a photo, but also how it is laid out. Using the tool, users can enter multiple search terms and drag them into different positions. The site will then display images that include those objects in similar positions.

For example, a user can enter the term “cat” and drag it to the top left, then enter the term “child” and drag it to the bottom right. The search engine would then return pictures with a cat and a child in those positions. The search tool even allows users to search for images with copy space, making it easier for businesses to find pictures with room for their existing media assets.

“What’s remarkable about this breakthrough is that we only trained our model to learn what things are, but our deep network learned how to represent where things are,” said Jon Oringer, founder and chief executive of Shutterstock. “For marketers, searching for an image with copy space using this tool will save a significant amount of time. We continue to innovate on this valuable search technology and invest in machine learning to improve the customer experience and provide more time for productivity and creativity.”

Shutterstock’s Composition Aware Search is currently in beta testing, but the tool is now available for anyone to try out. The company developed the search tool in-house using its own computer vision team.

Shutterstock is only one of several companies looking to AI to improve image search. Earlier this year, Pinterest unveiled an AI-powered camera app called Lens, which can recommend pins to users based on the pictures they take. Social media giant Facebook Inc. has also been developing new image search tools, such as the computer vision program it created that lets users search by image content.

Photo: Shutterstock