Shutterstock has been working on machine learning-based technology for different features, such as reverse image search and visually similar image search, and last week they unveiled a new feature that will allow users to search images based on their compositions and layouts.
This tool will let users to search for specific subjects by entering the related keywords, and then by simply changing the placement of the symbolic letters — which represents the keywords — in the "layout" section and Shutterstock will show the images based on users’ choice of composition.
While this may be useful for content creators for a fast and efficient way of finding the right images, it may also provide contributors more sales by bringing their lost images to light. In the traditional image search, the system usually brings thousands of results, sometimes including irrelevant images in hundreds of pages and this new feature will most likely change this positively.
Even though this feature is still in beta and it's patent pending, it looks like the company will continue using artificial intelligence and deep machine learning and implement these advantages to image search sooner than expected.
You can test the feature in beta on the Shutterstock website, and don’t forget to share your experience and thoughts in the comments section below.