Dreamstime already uses AI to tag photos with relevant metadata to improve the search results. The newest machine learning software isn’t going to replace actual photo editors, Dreamstime says, but will help make the process faster by labeling submissions and continue learning from the editor’s choices.
On the simplest level, the software can flag a photo of a person that wasn’t submitted with a model release and request the paperwork. The program can also recognize when a copyrighted logo is included in the image as well as other factors that trigger rejections. Flagging landmarks that can only be sold under an editorial license, marking duplicate submissions, and categorizing a submission as an illustration or a photograph also help speed up the process. The program then labels the images before they head to the actual photo editors.
But along with flagging potential problems, the system also recognizes a photograph’s subject and potential commercial value, Dreamstime says. As staff continues watching out for legal, aesthetic, and technical issues, the machine learning system will continue to improve the accuracy of the labels it assigns.
The company says the software will help photographers see a faster submission process, which in turn gets images selling faster. The system could cut review time by as much as 30 to 60 percent. In some cases, the computer creates an instant response.
“Constant innovation is a primary reason for our continued success in the stock image industry,” Serban Enache, the company’s CEO and co-founder, said in a press release. “With our AI solution, we have created a continually improving tool that will conduct instant image reviews, which will streamline our approval process. And we can layer on other types of metadata points such as customer behavior to build a truly intelligent solution that helps us improve the end-user experience.”
Dreamstime hopes to continue improving the program, with a goal of eventually factoring in sales data to better determine the type of images that are selling the most.
