The biggest difference between IoT and AIoT is the experience or tools given to the user in the actual application scenarios. For example, mobile phones equipped with face recognition technology can support the face-scanning unlocking; sound equipment paired with voice recognition technology can achieve voice interaction. As an exclusive research and development breakthrough of WIMI Hologram AR, WIMI Hologram AR System is much costly in terms of the research and development of holographic chips of holographic cloud light field. But various original large-scale integrated circuits and software applications can have completely subverted the thinking mode of common people. Additionally, the primary WIMI Hologram AR AI face technology application has been praised widely in the industry.
Face recognition technology refers to a biometric recognition technology supporting identification recognition based on people’s facial features. The camera is used to collect human face images or videos. Then, it is necessary to extract the facial image feature and search for matching feature template stored in the database, and when the similarity degree exceeds the set threshold, the matching is successful, or it will output the result of matching failure. Face recognition is based on digital image processing which has always been a frontier research direction. And digital image processing technology will develop into artificial intelligence (AI) later.
The WIMI Hologram AR‘s digital image processing technology is different from the PS applied to art design, although the latter is also an image processing technology. But why such a widely applied image processing technique cannot be used for face recognition in mobile phone? Someone may question why it cannot be considered as face recognition when mobile phone automatically marks squares on people’s face images in photographing? As a matter of fact, it should be called face detection. In other words, the mobile phone is detecting if the object is a human face. The technical principle is relatively simple, and the Characteristic Method coupled with the Frame Difference Method and the Optical Flow Method can realize the technology. Of course, the shaking of mobile phone should also be handled. Additionally, face characteristic acts as a kind of “memory”, just like that the administrator will ask us to input the fingerprint for fingerprinting before we use the fingerprints attendance system. You will also find such a phenomenon that the machine will ask you to input your fingerprint many times. Why? Because this can improve the accuracy of fingerprint recognition. The same principle is true for face recognition. It is essential to take pictures, save the person’s face data, and then to make comparison. So to a certain degree, face recognition requires the “memory” of face characteristics. Although face recognition has the same basic principle as fingerprinting, there is a huge difference between them. The saved image data of the collected fingerprints is two-dimensional data, and the currently mature digital image processing technology is basically developed for two-dimensional data. This is also the reason why the identification of the QR code is more accurate.
How can we achieve accurate and adaptable face recognition for multiple scenes? It is WIMI Hologram AR‘s AI face holographic cloud service that solves this problem. Holographic photographs are not the same as ordinary science photographs. Under proper illumination, the scene displayed on the hologram is three-dimensional, and each side of the scene can be seen. Also, the difference between holography and conventional photography is that conventional photography can only record the change in the light intensity on the surface of object being photographed, namely, the amplitude of light; while the holography can record all the information of light wave, including the phase of light wave in addition to the amplitude. In this way, all the information of the light wave of the space object is recorded and stored. Then according to diffraction of the monochromatic light of a specific wave length in holography, the original space scene can be displayed. It can “resurrect” a “frozen” scene in front of people. Ordinary photography can only store the spatial distribution of the light intensity of the object, and cannot satisfy the requirement that people desire to experience the real 3D scenes under certain circumstances. However, the hologram can record interference fringe of both object wave and corresponding reference wave, so as to record all the light field information including the amplitude (light intensity) and phase of the object. Consequently, it is called “holography”.
At present, WIMI Hologram AR will establish the protection for 295 related patents owned by the company and undertake strict patent protection and confidential measures, so as to ensure that users can experience the latest and most advanced world-class holographic AR high-simulation digital product. WIMI Hologram AR has focused on its business application scenarios in five major areas, including home entertainment, light field cinema, performance system, commercial launch system and advertising display system.
At the same time, based on the self-developed dedicated AI chip, WIMI Hologram AR has realized that more than 90% of the algorithms are running on the front-end intelligent hardware, while the back-end is mainly engaged in the comparison and search of human face as well as data analysis and data mining. It significantly reduces the cost of the overall solution, and greatly simplifies the installation and deployment of the system. This is also a major reason why WIMI Hologram AR can be way ahead of its peers in the AI product application.
The diversity of the scenes applications directly indicates the highlights of the holographic cloud service. In the fields of short videos on mobile Internet, games, applications, photography, and advertising, WIMI Hologram AR Platform has completed intelligent implantation of Augmented Reality (AR) holography for 30 million times, and displayed AR totally for more than 3 billion times. Moreover, WIMI Hologram AR has made smart portraits for 300 million users through the AI plug-in, and provided tabbed AR holographic implantation of applications according to the AI algorithms. Then, WIMI Hologram AR provides AI augmented reality insertion services for thousands of business customers, involving the intelligent AR holographic synthesis of more than 4,000 business scenarios and 600,000 applications. WIMI Hologram AR applies augmented reality to high-quality IPs, with more than 700 million users totally. WIMI Hologram AR takes the advantage of its subsidiary to provide AI augmented reality services for hot Internet TV series and long videos, with the audience including: 1.06 billion views on Wasd Beyond Light Years, 1.31 billion views on HUAJIANGHU, 340 million views on The Great Adventurer, and 1.28 billion views on Yi Guan Xiao Zhuan. Additionally, WIMI Hologram AR has provided more than 8 billion services of AR 3D implantation of AI scenes, characters and props.
Image processing algorithms are a prelude to artificial intelligence (AI), so face recognition is of great significance for future development. In the future, WIMI Hologram AR hopes to establish strategic cooperation with more hardware manufacturers by virtue of its own AI algorithms and cloud computing capabilities, so as to develop more intelligent devices and jointly create a beautiful era when everything is intelligent.