Intel demos 5G facial-recognition payment technology
origin : ZDNet
Date : 2017-12-07
Views : 0
In a showcase of several 5G-enabled technologies this week, Intel and Foxconn have demonstrated how facial-recognition capabilities can be used to make payments. Such "pay via face identification" technology could allow consumers to make payments for entertainment and retail services using Intel's Multi-access Edge Computing (MEC) 5G solution along with advanced facial-recognition technology and artificial intelligence (AI). "MEC and facial recognition systems used in smart retail can complete the payment authentication process within 0.03 seconds," Intel said. "This reduces the risk of personal information leakage and credit card fraud. Facial recognition is expected to be introduced to a variety of industries as the means of payment authentication, without the need of cash or credit cards." Intel and Foxconn used Asia Pacific Telecom's mobile network for the demonstrations. "Wireless cameras can connect to the internet via small cell radio access nodes offered by Asia Pacific Telecom and transmit the captured image to the facial-recognition engine to be processed on the local MEC platform," Foxconn Technology Group corporate executive VP and Asia Pacific Telecom chair Fang-Ming Lu said. "The low-latency, high-speed computing and assessment capability allows the system to deliver the computing result in real time to achieve premium customer experience." According to Intel, the MEC solution involves servers being installed at various locations, which are then monitored and managed via a centralised MEC Controller. Wired and wireless network access options are supported, which Intel said ensures flexibility. "The MEC solution is developed from the integration of Intel's Network Edge Virtualization (NEV) SDK with open-source software DPDK (Data Plane Development Kit) which together provide the 5G high-speed connectivity by delivering edge computing capability," the company added. Intel and Foxconn also demonstrated facial recognition use cases across residential and office access control and virtual shopping malls. Facial recognition access control for the office and home would involve using a 5G-backed high-speed authentication capability with blacklist and VIP alerts built in. Offices could track their employee attendance and working hours, Intel said. Using facial recognition for smart, virtual shopping, meanwhile, would personalise the experience for users by tracking their "historical shopping behaviour data" and serve them with an image of how a specific item of clothing would look on them. "Apparel boutiques can use MEC to quickly download 3D virtual apparel and accessories for consumers to try on via a virtual mirror," Intel described. "This not only makes the shopping experience more personal for the consumer, it reduces the amount of space required for inventory." Stores could also use the facial recognition data to serve targeted advertisements based on age, gender, and category. "The technology can be used in shopping malls to deliver indoor navigation, a 360-degree panoramic view, to store information, and for promotions," Intel added. Intel had last month similarly described how it would use "5G security" facial-recognition technology to transmit drone and camera footage to the control centre for security purposes during the 2018 Pyeongchang Olympic Winter Games. The company's focus on 5G technologies has seen it unveil a 5G modem suite and 5G mobile trial platform in recent months. The first modem from Intel's 5G New Radio (5G NR) modem suite, the XMM 8060, will be made commercially available in mid-2019. The XMM 8060 will have multi-modal functionality, meaning it will support non-standalone (NSA) and standalone (SA) configurations for 5G; millimetre-wave (mmWave) and sub-6GHz spectrum bands; and legacy mobile networks. Its third-generation 5G Mobile Trial Platform was unveiled in September, allowing for device development by supporting initial 5G NR specifications in live tests with partners by the end of this year, which will then be developed alongside 3GPP standards. It is powered by Intel's field-programmable gate array (FGPA) circuits and Core i7 processors, and supports the 600-900MHz, 3.3-4.2GHz, 4.4-4.9GHz, 5.1-5.9GHz, 28GHz, and 39GHz spectrum bands with a mobile interoperability solution for end-to-end 5G field testing. Intel in September also announced a self-learning neuromorphic AI test chip named Loihi, which is aimed at mimicking brain functions by learning from data gained from its environment. Applications could include the use of facial recognition via streetlight cameras for solving missing person reports; stoplights automatically adjusting to the flow of traffic; and robots gaining greater autonomy and efficiency, Intel said at the time.