Viewpoint-invariant face recognition based on view-based representation

Authors:
Jinyun Chung;Juho Lee;Hyun Jin Park;Hyun Seung Yang
Affiliations:
Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea;Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea;Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea;Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea
Venue:
ICIC'06 Proceedings of the 2006 international conference on Intelligent computing: Part II
Year:
2006

Citing 4
Cited 0

Unsupervised learning

Unsupervised learning
Face Recognition from Unfamiliar Views: Subspace Methods and Pose Dependency

FG '98 Proceedings of the 3rd. International Conference on Face & Gesture Recognition
Illumination-Based Image Synthesis: Creating Novel Images of Human Faces Under Differing Pose and Lighting

MVIEW '99 Proceedings of the IEEE Workshop on Multi-View Modeling & Analysis of Visual Scenes
Emergence of Phase- and Shift-Invariant Features by Decomposition of Natural Images into Independent Feature Subspaces

Neural Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we suggest a viewpoint-invariant face recognition model based on view-based representation. The suggested model has four stages: view-based representation, viewpoint classification, frontal face estimation and face recognition. For view-based representation, we obtained the feature space by using independent subspace analysis, the bases of which are grouped like the neurons in the brain's visual area. The viewpoint of a facial image can be easily classified by a single-layer perceptron due to viewdependent activation characteristic of the feature space. To estimate the independent subspace analysis representation of frontal face, a radial basis neural network learns to generalize the relation of the bases between two viewpoints. Face recognition relies on a normalized correlation for selecting the most similar frontal faces in a gallery. Through our face recognition experiment on XM2VTS [9], we obtained a face recognition rate of 89.33%.