Image provider Shutterstock Inc. has jumped on the machine learning bandwagon with the release of a new keyword suggestion tool on iOS that’s aimed at contributors.
The idea behind the new tool is simple – it’s tricky for mobile users who’ve uploaded a photo to spend time typing out relevant keywords for their images using such a small screen. So, Shutterstock is basically providing a helping hand by doing it for them, providing a bunch of keyword suggestions based on the concept the uploaded image portrays.
Shutterstock said the tool is being rolled out to mobile users first due to the difficulty of using a smaller screen, though it’s expected the tool will be made available to users on all platforms at a later date.
The new tool is based on Shutterstock’s recently announced computer vision tech, which scans uploaded images and locates visually similar ones, before generating “relevant and previously successful keywords” for that particular image. Uploaders can then just quickly select the keywords they wish to use, without needing to type them out, saving time and hassle. The tool relies on Shutterstock’s custom-made neural network that uses both pixel and metadata to identify images and come up with the most suitable keyword suggestions.
“We continue to invest in our Artificial Intelligence tools, which allow us to enhance the Shutterstock experience,” said Jon Oringer, Founder and CEO of Shutterstock, in a statement. "In this instance, we are helping our artists and photographers focus on the job they enjoy most. Keyword suggestions and other innovative tools built on our custom computer vision technology are instrumental in bringing efficiency and accuracy to the contributor experience, helping them earn more faster."
Shutterstock contributors can select up to 50 keywords for their images, giving themselves the best possible chance that their image will be found by site users.
The new tool is available to use with the Shutterstock Contributor iPhone application from today.