Intel launches RealSense face recognition ID on device

Intel today unveiled the latest addition to RealSense, its in-depth product range and tracking technologies designed to give the machines depth perception. Called RealSense ID, it is a solution on the device that combines an active depth sensor with a machine learning model to perform face verification.

Intel claims RealSense ID adapts to users as physical features such as facial hair and glasses change over time and work in different lighting conditions for people ‘with a wide range of heights or complexions’.

But numerous studies and VentureBeat’s own analyzes of public benchmark data have shown that face recognition algorithms are susceptible to different biases. One problem is that the datasets used to train the algorithms are white and masculine. IBM found that 81% of the people in the three collections with facial image most mentioned in academic studies have lighter skin. Academics have found that photographic technologies and techniques can also benefit lighter skin, including everything from separating film to low-contrast digital cameras. As a result, Amazon, IBM, Microsoft and other moratoriums on the sale of face recognition systems.

In response to this and other criticisms, Intel says that RealSense ID, which starts at $ 99 and launches in the first quarter of 2021, has built-in anti-spoofing technology to protect against fake access attempts using photos, videos or masks. The company also says it has a ‘one-to-one-million’ false acceptance rate and processes facial images locally, encrypts user data, and is activated by ‘user awareness’. In other words, it will not be verified unless requested by a pre-registered user.

Intel sees that RealSense ID is initially used for smart locks, access control, point of sale, ATMs and kiosks. The business eventually plans to expand into subcategories such as finance, healthcare and smart access control.

“We have been conducting extensive data collection of all ethnicities from Asia, Europe, Middle East Africa,” Intel’s head of product management and marketing, Joel Hagberg, told reporters during a briefing this week. “We were very careful to ensure that all ethnic groups were covered.”

Sagi Ben Moshe, corporate vice president, added: “We have invested a lot of money to build our own database … It was a huge data capture project that took us a lot of time, and it is which makes it unique. ” He and Hagberg did not disclose the size of the database or investigate the biases it may have, nor the method by which photos and topics were selected in the database.

RealSense ID follows the launch of Panorama Appliance by Amazon Web Services, a plug-in device from Amazon that connects to a network and identifies video streams from existing cameras. The Panorama device is equipped with computer vision models for manufacturing, retail, construction and other industries, and supports models built into SageMaker. It also integrates with AWS IoT services, including SiteWise, to send data for broader analysis. In addition to the Panorama Appliance, there is also the AWS Panorama SDK, which enables hardware vendors to build new cameras that drive computer vision to the edge. It works with slides designed for computer vision and deep learning from Nvidia and Ambarella, and Panorama compatible cameras work out of the box with AWS machine learning services.