Dahua Technology Ltd

Recently, the facial recognition technology team from Dahua Technology presented their results for Labeled Faces in the Wild (LFW), which is a worldwide authoritative facial recognition database test set. With a series of technical improvements, the Dahua FaceImage facial recognition system was not only ahead of Google, Facebook, Baidu, and Tencent, it also set a new LFW record and placed Dahua's facial recognition technology in a world leading position.

Improving facial recognition

LFW was established in 2007 by the University of Massachusetts and is used to evaluate the performance of facial recognition algorithms under unconstrained conditions. It is the most widely used evaluation set in the field of facial recognition. Up until now, a few dozen teams from around the globe have provided more than 80 test results. These teams come from companies including Google, Facebook, Microsoft Research Asia, Baidu, Tencent, SenseTime, Face++, Chinese University of Hong Kong, and other leading teams from industry and academia. Many of the top-ranking methods have played a crucial role in promoting the development of facial recognition technology.

In recent years, most of the
improvement in facial recognition
technology was achieved by the
development of deep learning

In recent years, most of the improvement in facial recognition technology was achieved by the development of deep learning. Deep learning is a machine learning approach that simulates the human neural system. The functions of Convolutional Neural Networks (a commonly used deep learning model) are closely related to network depth. However, since deep networks are difficult to optimise, the network depth for facial recognition methods of the past generally ranges from a few to a few dozen layers.

High accuracy

Dahua's facial recognition technology team has designed a network with a depth of over a hundred layers (this is currently the deepest network layer amongst facial recognition system that has been announced). This enables a new type of metric learning method which allows the similarity score to be higher for images of the same person. At the same time, it lowers the similarity score between images of two different people. Combined with a highly efficient online sampling technique, the rate of convergence can also be greatly increased. By training multiple models and using a non-linear multi-model integration technique, Dahua has achieved an accuracy of 99.78% for the LFW dataset.

Dahua's facial recognition technology team is part of the advanced technology research institute in their R&D centre

Dahua's facial recognition technology team is part of the advanced technology research institute in their R&D centre, where its abundant resources in deep learning technology, hardware, data and other aspects, supported the team to deliver high performance outcome. The team supports Dahua's facial recognition related products (face detection, face feature point location, face recognition, face attribute analysis, smile detection and so on) and provides the technology to ensure robust commercial implementation.

Implementation in real-life situations

Spokesperson for the Dahua facial recognition technology team, Professor Wang Haiyang indicated that achieving high accuracy in open data sets had motivated his team to work on the more challenging implementation in real-life situations. Dahua has been a leader in the security and video surveillance industry for many years, and accumulated a huge volume of video data. By utilising this data to adjust the learning model, the performance of the algorithm in real-life situations has been greatly improved.

Furthermore, the security application requires the recognition algorithms to respond quickly. The technical team used network pruning and a multi-model feature sharing technique to greatly reduce redundant operations so that the amount of calculations required for networks with over a hundred layers is similar to that for networks with only a few dozen layers.

At the latest G20 summit, Dahua's facial recognition technology was successfully applied to enhance the security level

Application areas

Dahua's facial recognition technology has already been applied to public security, finance, and other areas. At the latest G20 summit, which has been at the centre of attention, Dahua's facial recognition technology was successfully applied to enhance the security level. Dahua cameras were deployed in core G20 activity areas and in many traffic hubs. Real-time images captured were automatically matched with security blacklists in the back-end. When a high-risk person showed up in the video, the system will issue a warning message to command centre to dispatch response immediately.

This system played an important role during the G20 summit and assisted police in arresting a number of fugitives while the system was still in its commissioning phase. Customer oriented and open innovation have been the core value of Dahua Technology. Dahua's facial recognition technology reflects its commitment to provide customers with better technology and services to implement more sophisticated solutions at lower costs.

Download PDF version

Dahua Technology Ltd case studies

Case studies
Dahua Smart IoT Industrial Park uses integrated security technology for enhanced productivity

To grasp the current manufacturing trends and seize the Zeitgeist of Industry 4.0, a new smart IoT industrial park in Hangzhou has been put to use by Dahua Technology, a video surveillance solution provider. Dahua Smart (IoT) Industrial Park occupies in total 512 acres in Fuyang district of Hangzhou, about 20 minutes’ drive from Dahua headquarters, designed to host 6000 staff (by 2017, 4500 people have been working/living in the 262 acres of phase one area). With topnotch technologies, pe...

Case studies
Dahua delivers surveillance solution for Turkey's Sinan Erdem Dome

The Sinan Erdem Dome is the largest multi-purpose indoor venue in Turkey. Located in Istanbul, the dome has a seating capacity of up to 22,500, and hosts a number of events, including concerts, tennis matches, and basketball games. Strengthening stadium security Upon being chosen to host a number of games during the European Basketball Championships 2017, the chief European men’s international basketball competition held biannually, the Sinan Erdem Dome looked to strengthen their securit...

Case studies
Dahua secures G20 summit to ensure smooth running and safety of officials

The G20 Summit is an annual meeting of leaders from 20 major economies to discuss global issues. In 2016, China hosted its first-ever G20 forum in the south-eastern city of Hangzhou. Securing the leaders of multiple countries is no easy task, and would require many months of preparation by thousands of labourers in order to ensure the two-day forum, transportation, and cultural activities ran smoothly. China’s largest security project The G20 World Summit was one of the largest security...