Facial recognition cases occur frequently. Can multi-modal biometric verification become a new 'silver bullet”?

According to the facial recognition security survey released by the Ministry of Public Security in recent years, existing facial recognition technology has serious security risks. Criminals use publicly available user photos to copy and attack through electronic screens. , 2D paper printing attacks, 3D mask attacks, deep forgery attacks, adversarial sample attacks, ROM hijacking attacks, link hijacking attacks and other forms have broken through the technical defense lines of hundreds of well-known APPs, causing serious technical risks.

Therefore, how to improve the capabilities of biometric verification and anti-counterfeiting and solve existing security risks has become the core issue of biometric verification and anti-counterfeiting in the financial industry. In order to overcome this difficulty, Zhongguancun Kejin has designed a multi-modal biological verification and anti-counterfeiting algorithm fusion system. Based on the technical solution of audio and video fusion, by integrating sound technology and organically combining it with existing visual technology, it can greatly improve biological verification. Verification and anti-counterfeiting capabilities not only solve existing security risks, but also break down usage restrictions. The application of this technology can meet user-level anti-counterfeiting needs and can also bring more possibilities to intelligent supervision in various fields.

The multi-modal biometric verification and anti-counterfeiting algorithm fusion system realizes dimensionality reduction to combat electronic screen replica attacks, 2D paper printing attacks, 3D mask attacks, deep forgery attacks and other face attacks Security issues, and can form differentiated scenario-driven verification solutions, provide enterprise-level security protection capabilities, and meet user-level verification and anti-counterfeiting needs.

In the [T·Talk] technology sharing event on August 25, we specially invited Mr. Feng Yue, director of Zhongguancun Kejin AI Security Attack and Defense Laboratory, to be a guest in the live broadcast room , revealing to the general audience the technical principles and practical details of the Zhongguancun Kejin multi-modal biological verification and anti-counterfeiting algorithm integration system. Whether you are a practitioner in the algorithm industry or a developer keen on bio-anti-counterfeiting technology, I believe you can gain some unique technical experience and technical practices from this sharing.

Welcome everyone to participate in August 25th, 20:00 pm

[T·Talk] technology sharing event

Scan the QR code on the poster and make an appointment for the live broadcast immediately

The above is the detailed content of Facial recognition cases occur frequently. Can multi-modal biometric verification become a new 'silver bullet”?. For more information, please follow other related articles on the PHP Chinese website!