A CUHK research team led by Prof. Tang Xiaoou of the Department of Information Engineering and Prof. Wang Xiaogang of the Department of Electronic Engineering has built a novel facial recognition system with the highest accuracy in the world. Tested with thousands of picture sets, this recognition system could recognize faces at an accuracy rate of 99.15%, regardless of changes in lighting, make-up and camera angles, on Labeled Faces in the Wild, compared to the accuracy rate of 97.53% of humans.

'The key challenge of face recognition is to develop effective feature representations for reducing intra-personal variations while enlarging inter-personal differences,' said Prof. Wang Xiaogang. 'With Deep Learning, the system is provided with much more powerful tools to handle the two types of variations, thus significantly improving the accuracy of face recognition.'

Traditional video surveillance can only focus on a small number of objects in a very simple environment. With this new system, users can target thousands of objects in very complex environments, making it a powerful tool for law enforcement and security agencies to seek out individuals among a crowd of thousands. It also has many important applications, such as finding terrorists from surveillance videos, recognizing imposters at ATMs, and automatically tagging face images uploaded to social networking sites.