Skip to content

Latest commit

 

History

History
38 lines (35 loc) · 2.08 KB

README.md

File metadata and controls

38 lines (35 loc) · 2.08 KB

Carely

A real-time face recognition app developed to aid patients with early stage dementia for the NetApp WIN 2022.


Key Features


  • Recognize and identify friends and family in real-time
  • Add new faces via image or in real-time
  • Set reminders for medication
  • Get an alert for repeated calls ( an extra feature, since people with dementia tend to forget they've already made a call and proceed to call the person repeatedly )

Steps for face recognition


Face detection

  • Google ML Kit

Face Recognition

  • Convert Tensorflow implementation of MobileFaceNet model to Tflite

Face-recognition in real time

Set reminders for medication

Get an alert for repeated calls


Model used - FaceNet

  • Accuracy - 98.57
  • Precision - 100
  • Recall - 97.14
  • F1 Score - 98.55

What's Next?

  • Add a feature to mark contacts as favourites and display them on top.
  • Add multi-language support to increase accessibility.
  • Understand patient requirements and take feedback.