Low-power AI wearable recognises user’s face

Article By : KAIST

The K-Eye series can detect a face at first and then recognise it, and it is possible to maintain “Always-on” status with less than 1mW power consumption.

Researchers from the Korea Advanced Institute of Science and Technology (KAIST) have teamed up with start-up UX Factory to create a face recognition system, called K-Eye, using the South Korean institute's semiconductor chip—the CNN Processor (CNNP)—which runs artificial intelligence (AI) algorithms using low power.

The K-Eye series consists of two types, according to the researchers. The wearable type device can be used with a smartphone via Bluetooth, and it can operate for more than 24 hours with its internal battery. Users hanging K-Eye around their necks can conveniently check information about people by using their smartphone or smart watch, which connects K-Eye and allows users to access a database via their smart devices.

Meanwhile, a smartphone with K-EyeQ, the dongle type device, can recognise and share information about users at any time.

When recognising that an authorized user is looking at its screen, the smartphone automatically turns on without a passcode, fingerprint or iris authentication. Since it can distinguish whether an input face is coming from a saved photograph versus a real person, the smartphone cannot be tricked by the user’s photograph, according to the researchers.

Figure 1: Schematic diagram of K-Eye system. (Source: KAIST)

The K-Eye series carries other distinct features. It can detect a face at first and then recognise it, and it is possible to maintain “Always-on” status with less than 1mW power consumption. To accomplish this, the research team proposed two key technologies: an image sensor with “Always-on” face detection and the CNNP face recognition chip.

The first key technology, the “Always-on” image sensor, can determine if there is a face in its camera range. Then, it can capture frames and set the device to operate only when a face exists, reducing the standby power significantly. The face detection sensor combines analog and digital processing to reduce power consumption. With this approach, the analog processor, combined with the CMOS Image Sensor array, distinguishes the background area from the area likely to include a face, and the digital processor then detects the face only in the selected area.

The second key technology, CNNP, achieved incredibly low power consumption by optimising a convolutional neural network (CNN) in the areas of circuitry, architecture and algorithms. The on-chip memory integrated in CNNP is designed to enable data to be read in a vertical direction as well as in a horizontal direction, according to the researchers. It also has immense computational power with 1024 multipliers and accumulators operating in parallel and is capable of directly transferring the temporal results to each other without accessing to the external memory or on-chip communication network. Third, convolution calculations with a two-dimensional filter in the CNN algorithm are approximated into two sequential calculations of one-dimensional filters to achieve higher speeds and lower power consumption.

With these new technologies, CNNP achieved 97% high accuracy but consumed only 1/5000 power of the GPU. Face recognition can be performed with only 0.62mW of power consumption, and the chip can show higher performance than the GPU by using more power, researchers said.

The research team and UX Factory are preparing to commercialise the K-Eye series by the end of the year.