Face Recognition

Aim of this project is to provide an end-to-end face detection and recognition solution which utilizes network compressions and various data augmentation techniques to shrink models sizes to make overall solution capable to work on-the-edge devices as an online solution. We also aim to overcome detection & recognition problems due to dark skin and low-light environments.

Network compression for lighter and faster models. Real-time, on the edge solutions for detection and recognition. Data augmentation against dark skin and low light oriented problems.

1. Face Detection
2. Face Recognition
3. Dark Skin Recognition
4. Data Augmentation