Facebook Face Recognition

What is it?

It is a machine learning system that analyzes the pixels of a face in an image and assigns a unique template to it.

97% accurate according to FB

FB user turns on "face recognition"

Someone tags you in a photo

Face analysis

Template created

New photo uploaded

Comparison: faces

vs templates

Match

Tag suggestion

How does it work?

• Someone pretending to be you

• Identifying criminals 

• What if you’re caught in the background of a photo by happenstance?

Pros

Cons

• Finding potential terrorists

• No key or password that can be stolen or lost.

Safety and Security

Speed and accuracy

• High implementation costs

• Data storage and use of memory

• Changes in appearance

1. First, look at a picture and find all the faces in it                                                      

Challenges and Machine Learning Steps

Make it black and white. Pixels and vectors

2. Second, focus on each face and be able to understand that even if a face is turned in a weird direction or in bad lighting, it is still the same person.

68 landmarks and then rotate and scale                                           

3. Third, be able to pick out unique features of each face

Embedding (128 measurements)

4. Finally, compare the unique features of that face to all the other's in the database

any basic machine learning classification algorithm