Foap is company that provides a visual content platform for photographers to upload and sell their images. Foap's customers are businesses and creative agencies that can search and buy images from their app. They have a community of 3 million designers and host millions of photos and video assets.
Foap's business is critically dependent on their ability to accurately tag photographer uploaded images. They need to make it as easy as possible for businesses and agencies to find what they are looking to prevent customers from going elsewhere to buy their images. Since Foap receives thousands of uploaded photos every day, they needed a way to speed the tagging accuracy of large volumes of image uploads. Accuracy of image tagging was paramount for optimizing user find-ability.
Because Foap’s photos come from many individual sources, the tags and metadata attached are often inconsistent. For example, an image of a dog might be tagged “dog,” or it might be tagged “fuzzy waggy tail fur baby.” Foap needed a solution that could “see” each image and apply the appropriate tags accurately and consistently.
Foap turned to Clarifai for their visual search capabilities. With so much user generated content, Clarifai's deep learning platform was used to automatically categorize and label the large volumes of user generated content being uploaded. Using Clarifai's pre-trained content moderation model they were able to build a solution to moderate image content at scale. Clarifai's pre-trained moderation models were used to find and eliminate inappropriate content before it was posted to their website.
Foap’s founder, David Los, implemented Clarifai as a cornerstone of his product for its accuracy, flexibility, and affordability. With a team of five developers, David was able to get Clarifai’s core model up and running in only a couple hours and a few lines of code. By investing in Clarifai, David’s team saved 40 hours per week in manual keywording and moderation.
With Clarifai's content moderation model, they were able to easily identify and remove low quality and inappropriate pictures from their site. This pre-trained model saved hundreds of manual review hours weekly and freed up team members to spend their time onboarding new clients.