New program can now learn to identify blurred faces

September 19, 2016

A new type of technology is now able to remove blur from faces and instantly recognize the face behind it.

When you see a face that is blurred or pixelated, it usually means that the person does not want his face being public, especially if it involves security or other related incident. But a new algorithm can soon easily remove the blur on faces and recognize the person behind it.

One of the researchers, Vitaly Shmatikov, told Quartz it is using open-source machine learning software fo the algorithm to work. “We’re using this off-the-shelf, poor man’s approach. Just take a bunch of training data, throw some neural networks on it, throw standard image recognition algorithms on it, and even with this approach…we can obtain pretty good results,” he said in a report by Quartz.

It was stated though that the algorithm was developed to improve the vulnerabilities found on image encryption. By showing this, developers can create a better encryption system that could protect the privacy of people.

[Image credit: McPherson et al.]

by Victor Philip Ortiz
Tech Enthusiast and Movie Buff. When he’s not busy playing with the latest games, Victor usually spends his time collecting Blu-rays and building his own movie library. As the Online Editor for T3 Middle East, he develops and writes content for which includes reviews, features, and videos in addition to managing its social media and web content.