Communications of the ACM
Face Recognition System Achieves 99.15 Percent Accuracy
Researchers Xiaogang Wang (left) and Xiaoou Tang (right) say their face recognition system could recognize imposters at ATM machines.
Credit: The Chinese University of Hong Kong
Researchers at the Chinese University of Hong Kong (CUHK) say their novel facial-recognition system is the most accurate system in the world. Humans recognize faces at an accuracy rating of 97.53 percent on Labeled Faces in the Wild, according to a team led by professors Xiaoou Tang and Xiaogang Wang. However, CUHK's recognition system, which was tested using thousands of picture sets, can recognize faces at an accuracy of 99.15 percent, regardless of changes in lighting, make-up, and camera angles.
The researchers say their system marks the first time computing algorithms have reached human-level performance on face verification. "The key challenge of face recognition is to develop effective feature representations for reducing intra-personal variations while enlarging inter-personal differences," Wang says. "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."
The technology has applications in security, law enforcement, Internet, and entertainment, Wang says. For example, the system could help law enforcement and security agencies identify individuals among a crowd of thousands, because it can target a vast multitude of objects in very complex environments.
From The Chinese University of Hong Kong
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found