What is the difference between Face Detection and Face Recognition in iOS perspective

Question

May be this sounds like a stupid one, but i really curious to know that, what is the difference between "Face Detection and Face Recognition" in iOS perspective? And in what case or which kind of situation should i use the two of them. I am new in iOS and have never any previous tinkle down experience about iOS face Detection/Recognition related thing. I am going to make an application, where i have to detect user face (by camera, not after taken photo) with database picture collection. Please give you response if any, and please don't misunderstand my question. O:)

-Thanks a lot in advanced.

Mcchesney

I can't answer for what it means specifically from a iOS perspective but the difference between face recognition and detections lies between whats actually being done. Detection refers to merely finding the face in a given image whereas recognition means to actually map the detected face to a particular reference face from a database (and subsequently identifying the person). Applications for sole face detection without recognition involve camera focus systems where who the face belongs too doesn't matter. Recognition can then be used to tag the people after the shot is taken. – Wina

Thanks for your explanation. Now, i get it. :) it seems that i have to work with Face Recognition. – Mcchesney

Answer 1

iOS has face detection, but no face recognition. It can tell you where the faces are in an image but can't tell you who they are.

If you want to use the face detection, start with AVMetadataFaceObject or a tutorial like this one.

Exostosis

Thanks for posting your comment. It is really help me a lot. Have a nice day :) @nevan king – Mcchesney

Answer 2

In General:

Face Detection: Detect the face in the image. It searches general human face like segment in the whole image. Output may be one or more than one. The output will be a rectangle or rectangles on the faces in the image.[Paul viola method]

Face Recognition: Recognize input face from the already trained database with highest match score. A single face should be given as input, and the output will be a name, or class name or unknown face. [PCA, LDA]

Occupation