Case Study: Volumental achieves accurate shoe sizing and fewer returns with Sama

Accurate Data Labeling Powers the Volumental Shoe Sizing App

Volumental, the company behind a mobile foot scanning app that powers shoe recommendations for millions of shoppers, needed highly accurate computer vision data to improve 3D foot measurement. Existing AR frameworks weren’t precise enough, so they turned to Sama to help label datasets for their computer vision models with pixel-perfect accuracy and the ability to handle edge cases like shadows, low contrast, and occlusions.

Sama delivered high-quality annotations with an average Quality Score of 97.84% and maintained tight feedback loops so Volumental could refine its models and scale labeling as needed. With Sama’s support, Volumental improved the accuracy of its foot scans, helping create more personalized shopping experiences, reduce online returns, and drive more revenue for retailers and brands.

Volumental

Mikael Andersson

Senior Product Owner