a Person Add a Face to the Train the persongroup PersongroupID PersonName PersongroupName PersonID Image (stream / url) PersongroupName PersongroupName
1. Train a model with Custom Vision (CoreML) 2. Send the detected image to the Computer Vision API Scan object 90% confidence Read handwriting frame Classify Send to Vision API
Sample Format 16 bit Channels Mono Record the phrase Call the API (VerificationProfileId & Enrollment sentence as bytes) Handle response (result: Accept | confidence: High)