Wayvy
Face recognition and health data that never leave the phone.
Challenge
People losing weight watch the scale move faster than the mirror. Progress photos show the change but sit unsorted among thousands, and finding someone in their own library meant building face recognition without sending a face or a weight to any server.
Solution
Wayvy, Tapiki’s own iPhone app, pairs each weigh-in from Apple Health with a photo of the user from the same day. Face recognition runs entirely on the phone. Apple Vision finds and aligns faces, and a deep learning model converted to Core ML matches them against a few reference photos.
Health records and biometric data stay on the device. There is no account, no server, no analytics, and no third-party code, and recognition takes about 30 milliseconds per photo.
- Client
- Wayvy
- Sector
- Health & Science
- Role
- Product, design, and iOS development
- Built with
- Core ML, Apple Vision, SwiftUI, HealthKit, PhotoKit