omniture

Ping An Technology Facial Recognition Technology Tops the World in LFW Test Results

Ping An
2017-05-12 21:53 7365

Applied 200 Million Times in Over 100 Usage Scenarios

HONG KONG, May 12, 2017 /PRNewswire/ -- According to the latest published test results from Labeled Faces in the Wild (LFW), the internationally acclaimed facial recognition technology authority, Ping An Technology Facial Recognition Technology was given a face recognition ratio of 98% and the lowest error range, a result that was superior to that of many renowned domestic and foreign rivals. It is also the best results LFW has published and marked a milestone in the success of Ping An Tech AI Lab. Moreover, the Certification Center for the Third Research Institute of the Ministry of Public Security (CSPHS) certified that Ping An Technology Facial Recognition Technology equally received an accuracy rate of 99.84% in test results, and CSPHS has also granted Ping An Technology Facial Recognition Technology a test report certificate.

At present, Ping An Technology Facial Recognition Technology has been applied to more than 100 usage scenarios, involving close to a hundred famous organizations and yielding an application rate of over 200 million times. Chen Liming, General Manager of Ping An Tech, said: "Artificial Intelligence (AI) is the way of the future when it comes to social development. Face recognition technology, a core segment of the AI spectrum, is getting more attention and deployment of usage. With the accumulation of service data, the endorsement of authorities at home and abroad and academic institutions, Ping An Technology Facial Recognition Technology has already attained an internationally leading position.

LFW (Labeled Faces in the Wild) released its latest test results
LFW (Labeled Faces in the Wild) released its latest test results

In order to accurately recognize the face identity, Ping An Tech's AI Lab makes use of various technical means for image comparison based on the facial features of humans. In the past three years, based on the accumulated user data by Ping An Group's commercial scenarios, Ping An Tech's AI Lab has continued to carry out thorough study and training on its own with a view to achieving high-accuracy identification and the highest levels in security requirements and stability of use.

Through repeated testing and usage, Ping An Technology Facial Recognition Technology has gradually embraced more and more application scenarios in the business units. In the financial application scenarios, Ping An Tech has successfully launched dozens of projects since 2015, covering credit approval, credit card batch comparison, banking business identity verification, internet account registration, account log in, insurance and underwriting. At present, Shenzhen social security wallet and enterprises including Ping An Puhui, Ping An Bank and Bank of Jiaxing are using Ping An Technology Facial Recognition Technology.

In addition, in terms of the application scenarios in government agencies and institutions, Shenzhen airport and Ping An Technology Facial Recognition Technology achieved a strategic cooperation and have started implementing projects. The Shenzhen Housing Authority and high-speed train identity checking have developed pilot projects and the Beijing-Shanghai High Speed train link is about to formally launch its online application. In Shenzhen, Fujian and Tianjin, the Ministry of Human Resources and Social Security is carrying out pilot testing of facial recognition for its identification documents. The work attendance, door access control of Ping An are also undergoing pilot testing. Ping An Tech is actively expanding into overseas markets and many international financial institutions have shown interest in forming cooperation.

Ping An Tech is proactively developing technologies related to AI in various aspects. Besides face recognition technology, it is committed to making work and life simpler and more intelligent through different kinds of research. Ping An Tech's audio navigation system covers all of Ping An Group's telemarketing business and is optimizing the experience of three APP applications -- Securities, Mobile Bank and Happy Ping An. Voice print analysis accesses the quality of customer conversation to gauge users' mood and emotions.

About Ping An Tech

Founded in 2008, Ping An Tech is a wholly-owned subsidiary of Ping An Group. It has more than 4,000 IT technologists and IT management professionals. As the high-tech core business of Ping An Group, Ping An Tech adheres to the roadmap of "Technology Leads Finance" and continues to develop mobile web, cloud and big data applications. It establishes a leading IT organization, strengthens the foundation to support core business, and takes lead and collaborates with all business units of Ping An to innovate and explore new internet-based business models.

Photo - https://photos.prnasia.com/prnh/20170512/1850472-1

Source: Ping An
collection