Facial Recognition has been gaining acceptance in recent times because of the benefits that it delivers over other surveillance techniques. The technology is defined as a method of recognising a human face through hardware like video cameras. Today the technology is being used by the retailers to pass on benefits like personalised experience when shopping in-store to the customers. Moreover, it is also being used to understand the customers and increase sales through features such as target advertisement and by allowing hassle-free checkout without the need for any card or cash.
All this is possible as over the time, the technology has only got better. Independent tests by the US National Institute of Standards and Technology (NIST) found that between 2014 and 2018, facial recognition systems got 20 times better at finding a match in a database of 12 million portrait photos. The failure rate fell from 4 per cent to 0.2 per cent over the period, a massive gain in accuracy that was largely due to deep neural networks. The network has driven ‘an industrial revolution’ in facial recognition technology.
How does the technology work
Facial Recognition is about facial geometry, which is the distance between eyes and depth of each socket like nose width and jawline length. All of these comprise what is known as facial signature.
The technology varies according to the need but the basic of it remains the same. The system captures the photo or video of the face. Then the facial recognition software reads the geometry of the face’s key factors including the distance between eyes and distance from forehead to chin. The software identifies facial landmarks (point of interest in an image); one system can identify almost 68 of them that are key to distinguishing a face (facial signature). This facial signature is then compared to a database of known faces and confirmation is made thereafter.
Facial recognition application by apparel retailers
Jack & Jones and Vero Moda, two brands owned by Danish fashion retailer Bestseller, opened smart stores in Shenzhen and Guangzhou, China. The stores were powered with facial recognition technology provided by Tencent’s YouTu Lab, one of the most advanced AI technology providers in China. The technology allows to pass on personalised recommendations to the customers and enables making payment without using any cash or card.
After completing the facial recognition registration process in-store, the shoppers immediately become members of the ‘AI Club’ powered by WeChat Pay. When they are done with shopping, they can complete their payment at the exit via WeChat Pay using their faces. The technology can also give personalised recommendations of Bestseller clothes and accessories that they might to try on virtually.
Ruti, a brand that focuses on apparel for women of 35 years and older, deployed facial recognition technology to offer personalised recommendations to the shoppers based on their preferences like sizes, likes and dislikes.
The technology works by scanning and taking photo of the customer’s face entering the store using the installed cameras. On customers’ approval, those photos (as well as items related to a purchase if one made) are stored in the retailer’s CRM system.
It can identify the repeat customers, tracing his/her shopping history and customer profile within seconds. Moreover, the technology can automatically pull five to six items from the most recent Ruti collection, along with pictures of those items and share recommendation sizes for the sales associate to pull, which help ensure the customer will have an easy try-on experience in the dressing room.
Moreover, facial recognition continuously builds customers’ profiles, including past purchases, items they have tried on and other attributes. Also, the feedback from the customers help influence the design process of the upcoming collection.
Technology providers for facial recognition
FaceX
Based in India, FaceX is a provider of facial recognition technology helping retailers with features like facial detection, facial landmark, face recognition, face tracking and spoof detection.
One prominent feature that the company provides is customer counter that can tell the number of customers entering the store throughout the day. The technology uses Smart CCTV cameras linked to the FaceX face detection API to process the information. It can brief the brands and retailers on information like gender-age identification along with the time and date of their visit to help retailers have better understanding of their store customers.
Moreover, the technology is also addressing the most requested demand of the clients that is analysing their review about a particular product. These smart cameras installed all around the store process information to tell how long each customer looked or glanced at a particular product using face recognition and face analytics. The technology also greets the customers that enter the retail store to pass on personalised touch to them. Besides, the technology can store certain attributes like the customers’ purchase history, invoice etc., with their face ID, creating customer database with their consent. When a customer revisits the store, the technology scans the incoming customer, identifies the repeat customer and the store employees are notified of their presence with the name stored in the database. “Facial recognition has a lot of potential not just in retail but in every industry possible. Initially we started testing the platform that was already existing and realised that our accuracy is much higher than the ones already in the market,” said Adarsh Mahajan, Business Development Manager, FaceX.
Using FaceX, the retailers and brands will also be able to pass on targeted advertisement to the customers. The FaceX API, when linked to the camera placed near the advertisement screen, scans each customer that is passing near by. The gender and age of every customer are detected by the FaceX system using face analytics and gender-age identification. The information is then processed by the system and an ad meant to target the respective age-gender group is displayed on the screen. The customer responds to the ad and can now be prompted by an on-screen message to visit that particular aisle number.
SightCorp
Started in 2013, this USA-based company specialises in easy-to-use face analysis and facial recognition technology. The facial recognition technology by the company is 100 per cent deep learning that provides greater accuracy.
The company provides attributes including age recognition, emotion recognition, face detection, face tracking, face detection and face analytics. The facial recognition technology by the company is trained to recognise, compare and verify faces in less than one second. It can identify and authenticate people. The patented proprietary AI software solution that uses computer vision, deep learning and machine learning can recognise the face in images, videos and real-life environment. The technology by the company combines digital ID verification and biometric facial recognition to identify the person.
The technology allows retailers to establish customer identity, helping them identify their customers and repeat customers. The data stored in the system would help identify the likes and dislikes of the customers, helping provide a personalised experience to the customers, thus increasing the client experience
