Picking faces out of a crowd is something humans are hardwired to do, but training computers to act in the same way is much more difficult.
There have been various breakthroughs in this field in recent months, but the latest could be the most significant yet.
Researchers from Yahoo Labs and Stanford University have developed an algorithm that can identify faces from various different angles, when part of the face is hidden and even upside down.
At the moment, the so-called Deep Dense Face Detector doesn’t recognise who the individual faces belong to, just that there is a face.
But the technology has the potential to be trained in this way.
The algorithm was built by Sachin Farfade and Mohammad Saberian at Yahoo Labs in California and Li-Jia Li at Stanford University.
It built on the Viola-Jones algorithm which spots front-facing people in images by picking out key facial features such as a vertical nose and shadows around the eyes.
By collecting these markers, the algorithm is able to determine if an image contains a face or not.
But, this did not account for faces that have been obscured, are looking in various directions, or were upside down.
With this in mind, Mr Farfade and his team used a form of machine learning known as a deep convolutional neural network.
This involves training a computer to recognise elements of images from a database using various layers.
Google used a similar technique for its recent GoogLeNet classification algorithm that can identify images within images, such as a hat on the head of a dog sat on a bench.
Mr Farfade trained his algorithm using a database of 200,000 images featuring faces shown at various angles and orientations, plus 20 million images that didn’t contain faces.
In their paper, the researchers said: ‘In this paper we propose a method based on deep learning, called Deep Dense Face Detector.’
‘It has minimal complexity…and can get similar or better performance [than other systems] while it does not require annotation or information about facial landmarks.’
And, the team said the technology could be improved following further training.
The Latest on: Facial recognition
via Google News
The Latest on: Facial recognition
- Speech recognition systems have higher error rates for black people, study findson March 23, 2020 at 6:51 pm
This is far from the first time we've seen evidence of bias embedded in the supposed technology of tomorrow. In December of last year, a federal study once again confirmed that facial-recognition tech ...
- Wolfcom Embraces Body Cam Face Recognition Despite Concernson March 23, 2020 at 3:30 pm
Even as it becomes ubiquitous in consumer tech, facial recognition has been increasingly verboten in the marketplace for police body cameras. Last June, Axon, the nation’s largest body-cam provider, ...
- There Is a Racial Divide in Speech-Recognition Systems, Researchers Sayon March 23, 2020 at 12:07 pm
Technology from Amazon, Apple, Google, IBM and Microsoft misidentified 35 percent of words from people who were black. Whites fared much better. SAN FRANCISCO — With an iPhone, you can dictate a text ...
- New Venture Shines Spotlight on Facial Recognition Collection and Databaseson March 20, 2020 at 8:00 am
Facial recognition software, and the databases that the images end up in, remain controversial. At the time of writing, a petition to stop the Met Police using facial recognition surveillance is a few ...
- Is ICE’s Use of a Maryland Facial Recognition Database Lawful?on March 19, 2020 at 5:00 am
The Washington Post reported recently that U.S. Immigration and Customs Enforcement (ICE) officials have accessed, without obtaining judicial process, a Maryland facial recognition database that ...
- Facebook’s own facial recognition app could show employees connections to their friend liston March 18, 2020 at 10:59 am
Facebook built a facial recognition tool that could be used by employees to scan faces of their friends who had certain features activated on the social media giant, according to a new report.
- How China built facial recognition for people wearing maskson March 18, 2020 at 6:45 am
Hanwang, the facial-recognition company that has placed 2 million of its cameras at entrance gates across the world, started preparing for the coronavirus in early January. Huang Lei, the company’s ...
- A Chinese firm claims it has built facial recognition tech that identifies people wearing masks with 95% accuracyon March 18, 2020 at 6:45 am
People have foiled facial recognition technologies by wearing masks during the coronavirus outbreak.
- San Diego police ban Clearview AI's facial recognition toolon March 16, 2020 at 7:13 pm
The backlash to Clearview AI's facial recognition tool is extending beyond tech companies and civil liberties groups. San Diego's police department and district attorney's office have confirmed that ...
- Washington state passes landmark facial recognition bill, reining in government use of AIon March 13, 2020 at 6:10 pm
The Washington state legislature passed a bill establishing new guardrails on government use of facial recognition software. The bill cleared both chambers of the state legislature Thursday, hours ...
via Bing News