Researchers have built a facial recognition system that outperforms human facial recognition abilities, recognizing faces at an accuracy of 99.15 percent compared to 97.53 percent. Developed by a Chinese University of Hong Kong (CUHK) team led by Professor Tang Xiaoou and Assistant Professor Wang Xiaogang, the novel
Researchers have built a facial recognition system that outperforms human facial recognition abilities, recognizing faces at an accuracy of 99.15 percent compared to 97.53 percent.
Developed by a Chinese University of Hong Kong (CUHK) team led by Professor Tang Xiaoou and Assistant Professor Wang Xiaogang, the novel facial recognition system can accurately identify faces regardless of changes in lighting, make-up and camera angles. This is the first time computing algorithms have reached human face verification performance on this dataset.
“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. “With deep learning, the system is provided much more powerful tools to handle the two types of variations and significantly improves the accuracy of face recognition. This technology has numerous important applications in security, law enforcement, Internet and entertainment.”
The system could help law enforcement and security agencies to seek out individuals among a crowd of thousands. Traditional video surveillance can only focus on a small number of objects in a very simple environment. With the new system the users can target thousands of objects in very complex environments.
Deep learning is the biggest breakthrough in artificial intelligence (AI) in recent years. It simulates human brain's behaviors by training large scale neural networks from big data based on intensive graphics processing unit computing.
Face recognition is one of the most important grand challenges in computer vision and AI. Scientifically, this is also an important benchmark on whether AI can reach the level of human intelligence or even surpass it. The research team's achievement is strong evidence that deep learning makes AI possible. It opens the door to many important applications, such as finding terrorists from surveillance videos, recognizing impostors at ATM machines, and automatically tagging face images uploaded to social networking sites.