Facebooks new face recognition System - Deep Face

 
We have seen face recognition technology in several Sci-Fi movies and it’s not new in the market. Even with the technology advancement, it is becoming cheaper, faster and easily available. A new dimension of possibilities has popped up with Facebook’s new face recognition system, "DeepFace". This technology can be boon for the security agencies, however, people anonymity won’t be possible.

This new technology has the ability to even predict who a user is trying to tag in a photo, since tags are not so accurate.This technology, they said, will operate with "near human accuracy" so that user wont have to tag photo´s manually in the future.

It actually maps 3D facial features, creates a colorless model, and then narrows the specific characterization. The Facebook API group said that its accuracy method scored 97.25%, which is closely to human accuracy score of 97.5%.

The researchers said in the released report that, “In modern face recognition, the conventional pipeline consists of four stages: detect => align => represent => classify. We revisit both the alignment step and the representation step by employing explicit 3D face modeling in order to apply a piecewise affine transformation, and derive a face representation from a nine-layer deep neural network.”









To help them develop this technology, the company looked at 4.4 million tagged faces from the 4,030 users in its network to test the system. This will help it to better recognize the specific features of each user.

“Thus we trained it on the largest facial dataset to-date, an identity labeled dataset of four million facial images belonging to more than 4,000 identities, where each identity has an average of over a thousand samples. The learned representations coupling the accurate model-based alignment with the large facial database generalize remarkably well to faces in unconstrained environments, even with a simple classifier.”the researchers said.

The company aims to bridge the gap between computer and human accuracy. However, we still have to see how the technology will run to strike the balance on privacy issues.

by: Ezekiel.T.Ogidan