Face recognition technology is increasingly innovative and breakthrough, and there are many cross-industry application projects. But at present, technology still can't keep up with the changes of society and market. For example, the outbreak of COVID-19 this year has caused a large number of domestic face products to be unable to scan and identify after wearing masks, and the algorithm updates of major manufacturers have not kept up. However, it also reminds us that in the face of future uncertainty, technology cannot be static and needs constant innovation and breakthrough.
Face recognition technology is still immature, and there are still many loopholes. According to relevant industry insiders, face recognition technology is overestimated and far less accurate than biometric technologies such as fingerprints or irises. At present, there are two major risk problems in face recognition that are difficult to solve. The first risk is repeatability. People are exposed every day, and it is possible to get a person's five senses by taking photos and copying them. Another risk is instability. Heavy makeup, allergies, injuries, facial plastic surgery, etc. It will cause great changes in the five senses, affect the accuracy of face recognition, and even be unrecognizable.
With the development of helmet simulation, holographic projection, face tracking and other technologies, the cost of future face recognition attacks will be greatly reduced, which will lead to a large number of hacker attacks. Criminals will destroy the system by forging face recognition and then steal confidential information. This will lead to frequent data leakage incidents and bring severe challenges to data security. ? Face recognition? Just like a key to open sesame seeds, open the system, enter the application, or use it? Brush your face? As a payment method. In this process, a large number of users? Face? Information is collected and stored. Not only Apple, but also other commercial organizations and related organizations? Legal? Get people's personal information. This is a huge individual in itself? Biometrics? Specimen bank.
No? Face? The collection will be leaked, resold by people with ulterior motives, or used to defraud interests, but has the facial technology really matured? Privacy? Tan times. The user's account information and facial features are leaked at the same time. Facial recognition? The technical integration of personal property has become a new trend. Therefore, it is not uncommon for individuals to expose fraud. It can be seen that the limitations brought by authentication technology are more and more obvious, and the integration of various security mechanisms has become a new trend.
First, strengthen research and development capabilities to improve the accuracy of in-vivo detection represented by face recognition; Second, pay attention to distinguish the use scenarios of identity recognition. When using advanced security scenarios such as privacy and payment, biometrics combines various factors, and at the same time, efforts are made in many aspects to improve the security threshold and ensure the safety of users.