Brush Face Carnival: iPhoneX does not necessarily achieve its own achievements but has achieved face recognition

(Original title: Face era, face recognition for entrepreneurs to break through the road)

On September 26th, an examinee entered a simulated examination room after confirming his identity through a face recognition system in a middle school in Beijing. Beijing News reporter Wu Jiangshe

Apple can always cause technological changes in the industry.

In an Apple conference in early September, “face recognition” was once again fired. The full-screen iPhone adopts Face Camp D technology, allowing the Chinese to fall into a “blush”.

“Brushing the face to collect the Provident Fund”, “face brushing and attendance”, “face security”... More and more daily behaviors have been crowned with “brushing the face” and even taken in the public toilets in Beijing’s Ritan Park. The toilet paper must brush his face.

With the combination of face recognition technology and government affairs, the future will have to prove that you are yours, and you may not need to cover more than a dozen chapters before and after, just swipe your face.

The people’s keen interest in face recognition is also reflected in the financing of companies in related fields: Face++, a face recognition cloud service platform, and Shangtang Technology, which focuses on building artificial intelligence vision engines, have received US$100 million in Series C financing and 410 million yuan respectively. Dollar B's round of financing is among the "unicorns" in the field of artificial intelligence.

The tipping point of face recognition technology into commercial applications seems to have arrived. iPhone X may not be able to accomplish itself, but it has achieved face recognition.

Three face recognition algorithms

What is face recognition? If the next definition is to be given to face recognition, it is a technique that uses human biometrics to differentiate individuals. It generally includes three steps: image acquisition, feature location, identity verification, and search. In simple terms, face recognition is to extract key points of facial features from the image, such as bone features, eyebrow height, etc., by comparing the output results.

Although the iPhone's Face Camp D has excited entrepreneurs in the field of computer vision, this face recognition is not face recognition. Apple's Face Camp D did not use the ordinary camera, but adopted the infrared active recognition technology, so that it can achieve three-dimensional stereoscopic recognition and enhance the safety factor.

In a specific application scenario, face recognition can be roughly divided into three types: 1:1, 1:N, and N:N.

The 1:1 level of face recognition achieves the first level of "prove that you are you." Literally, it can be seen that the 1:1 user uploads personal photos in advance and stores them in the system. In each verification, offline photographs are compared with photo information stored in the system to determine whether “you are not you”.

For example, when we passed the security check at the station, the teller took your ID card and compared it with yourself to prove that you were not on the ID card. This scenario is a 1:1 scenario. Phone unlock, face payment, online ticket purchase, hospital registration, government Huimin project, as well as various securities accounts, telecommunications account opening, Internet finance account opening ... ... This is a 1:1 face recognition application scenario. Compared with other methods, the accuracy of 1:1 recognition is high, and the requirements for computational power are relatively low.

The 1:N face recognition algorithm is mainly used for face retrieval, "to prove who you are." Unlike the 1:1 comparison, 1:N requires a photo to be compared with a large number of photos in the system, and multiple comparison results are arranged based on similarity. The result ranked first is not necessarily accurate.

The 1:N face recognition algorithm is mainly applied in the field of security, such as the investigation of criminal suspects and finding lost children. Yun Tian Li Fei, a start-up focused on dynamic portrait recognition, started cooperation with Shenzhen Longgang District Police in 2015 to build “Deep Vision” systems at local subway stations, railway stations, Chengzhong Village, and Shangchao. A few months after being on the line, the police assisted the police in successfully breaking two murders.

The application of 1:N face recognition in the security field is characterized by dynamic and non-matching. The so-called dynamic, that is, the system recognizes not the picture, but the video captured by the camera. Non-matching means that the recognition object does not need to perceive the position of the camera and cooperate with the recognition work, and the recognition object is in a passive state. This also means that the convenience of collection also means that your whereabouts have been completely exposed.

Compared with 1:1 recognition, the use of location, environment, light, collection angle and even glass reflection will affect the accuracy of 1:N recognition, so 1:N is relatively more challenging.

As for N:N face recognition, it is actually equivalent to performing multiple 1:N identifications at the same time for "proving who is who."

Face recognition technology is not perfect yet

As far as basic research behind face recognition technology is concerned, China, Europe and the United States are almost at the same level. But when it comes to commercial applications, China is absolutely ahead.

“Face recognition is not a very high-end technology at present. China’s big companies pay more attention to the immediate interests. Face recognition is the fastest and best way for them to realize artificial intelligence.” Body recognition technology at Beijing University of Aeronautics and Astronautics Experts in the field, Leng Rong, believe that large companies such as Google did not pursue face recognition because they had a longer-term plan.

The entrepreneurs in the field of artificial intelligence in China seem to love to get together.

From traditional Internet giants such as Alibaba, Baidu, Tencent, and Jingdong, to unicorns such as Face++ and Shang Tang Technology, to entrants who just entered the office, face recognition is not only a patent of a large company, but also a server. Face-to-face contend with mobile face recognition.

“The threshold for face recognition is very low, but it is not simple to do it in real-life scenarios.” Zhang Quanling, the founding partner of Zini Fund, represents the views of many investors in the industry.

In the TV program "Most Powerful Brain," Baidu's robot based on deep learning has basically demonstrated the current level of development of face recognition technology - it can easily extract thousands of feature points on the face of a person and use a large number of pictures for depth. Learning and training, and the preprocessing function of the face image, can effectively match the face image in 1 to 2 seconds.

At the same time, the small degree also exposed some of the difficult problems that face recognition needs to break through: Poor lighting conditions, different angles, blurred or distorted information can all cause recognition errors. In addition, face masks, hats, beards, hairstyles, and facelifts Or PS, etc. will interfere with recognition.

This year's 3-15th party exposed technical flaws in face recognition. The presenter quickly processed the 3D face model exactly the same as the viewer himself with a photo of the audience. The moderator puts on the audience's 3D face model, points at the camera of the mobile phone, completes actions such as blinking, turning, and smiling in accordance with the instructions of the APP, successfully tricks the system, and successfully completes the live detection and certification.

Qiu Xuejun, a visual analysis expert at 360 Artificial Intelligence Research Institute, said in an interview that at this stage, face authentication technology cannot be fully matured in all scenes. Although face recognition accuracy is high, most face recognition The system does not pay enough attention to living body detection. The algorithm used is relatively simple and it is not difficult to crack.

The scene is the key to the breakthrough of the entrepreneurial team

In addition to technology that is not yet fully mature, technology startups often fall into a myth: Only technical heroes. "Now most of the fledgling face recognition companies still stay at the level where my technology is better than others." Jiang Jiang, director of investment at PreAngel, said that these companies still do not think clearly about the scene of commercialization.

Zhang Hongjiang, a partner of source code capital investment, once said: “If you say that this company has only an algorithm, only a few cattle, no data or hard to get data, there is no application scenario, such companies do not do much.”

Of course, many companies also emphasize data, but this is not an absolute threshold. "Assuming that the existing technology has reached 92 points, if the new technology can reach 94 points, the difference between the user's feelings is not so strong, the significance is not significant." In the view of Wang Jun, Yun Tian Lifei Marketing Director, the technology is always Iteratively updated, but when the technology can not be subversive, the value of technology will be greatly reduced.

Of course, this does not deny the significance of technology. Only when unicorns have appeared on the track, technology has become very difficult to become the key for new companies.

"We are now focusing on start-up face recognition companies, seeing less pure technology, more concerned about companies that can solve practical problems." Jiang Wei said that commercialization capability is a breakthrough point for emerging companies. AI startup teams with equal technical and business abilities are more likely to be viewed by capital.

Whether it is from scene resources, data acquisition, or capital strength, the giants have the unparalleled advantages of the entrepreneurial team. After the giants enter the game, what should a startup company do?

“The foundation for the survival of a startup company is the deep service and penetration of the industry.” An investor who declined to be named said that the startup company will not be easily replaced by the innovation of the giant. On the contrary, the huge investment of the giant and basic innovation will promote Opportunities for the new sector emerged. For example, it was after Apple’s efforts to promote Apple’s mobile phone and Google’s vigorous promotion of the Android system that the development of mobile Internet really broke out. New opportunities such as Uber and Didi emerged. The appearance of Face Camp D is not a new opportunity for face recognition entrepreneurial teams.

However, it is still necessary to look at the scene if face recognition can be applied maturely. The above investors said: “I am more optimistic about some scenarios that do not require a high level of effect, such as consumption and entertainment, and are more cautious in areas such as financial payments where the accuracy rate is extremely high.”

This view coincides with Wang Jun, director of marketing for Yuntian Lifei. “The former GE chairman Jack Welch said in the book “The Essence of Business” that the value of the product itself is more important. This world is not the most cutting-edge technology and it has the most commercial value, but I think It is the easiest technology to replicate on a large scale, so I am optimistic about the application of face recognition in areas where safety requirements are not so high."

Beijing News reporter Cai Haoshuang

Solid-state capacitors / Motor starting capacitors

Solid - state capacitors are all called: solid - state Aluminum Electrolytic Capacitors.It with the ordinary capacitance (that is, the liquid aluminum electrolytic capacitors) the biggest difference is that use different dielectric material, liquid aluminum capacitor dielectric material as the electrolyte, and solid-state capacitor dielectric material is conductive polymer materials.Solid-state Capacitors / Motor Starting Capacitors

Solid-state capacitors / Motor starting capacitors,Solid-State Capacitors,Solid-State Small Size Capacitors,Solid-State Low Impedance Capacitors,Long Life Solid-State Capacitors

YANGZHOU POSITIONING TECH CO., LTD. , https://www.cnpositioning.com