Example-Based Learning for View-Based Human Face Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Face recognition: a convolutional neural-network approach
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Accurate face recognition is very hard to achieve but very important in practical applications. We propose a neural network, based on recognition by-recall paradigm, to improve the performance. We observe that human has remarkable capability in verifying if two photographs are from the same person or not, thus we devise a neural network to simulate the capability. We use eigen-face features to train the network and adopt a training method of the neural network to tolerate the errors in eye locations. The result shows that the verification performance of the system is very promising.