Grassmann Registration Manifolds for Face Recognition

Authors:
Yui Man Lui;J. Ross Beveridge
Affiliations:
Department of Computer Science, Colorado State University, Fort Collins, USA CO 80523;Department of Computer Science, Colorado State University, Fort Collins, USA CO 80523
Venue:
ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part II
Year:
2008

Citing 0
Cited 9

Aligning spatio-temporal signals on a special manifold

ECCV'10 Proceedings of the 11th European conference on Computer vision: Part V
Nearest-neighbor search algorithms on non-Euclidean manifolds for computer vision applications

Proceedings of the Seventh Indian Conference on Computer Vision, Graphics and Image Processing
Unfolding a face: from singular to manifold

ACCV'09 Proceedings of the 9th Asian conference on Computer Vision - Volume Part III
On advances in differential-geometric approaches for 2D and 3D shape analyses and activity recognition

Image and Vision Computing
Advances in matrix manifolds for computer vision

Image and Vision Computing
Age invariant face verification with relative craniofacial growth model

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part VI
A Gabor-block-based kernel discriminative common vector approach using cosine kernels for human face recognition

Computational Intelligence and Neuroscience
Human gesture recognition on product manifolds

The Journal of Machine Learning Research
Introduction to face recognition and evaluation of algorithm performance

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

Motivated by image perturbation and the geometry of manifolds, we present a novel method combining these two elements. First, we form a tangent space from a set of perturbed images and observe that the tangent space admits a vector space structure. Second, we embed the approximated tangent spaces on a Grassmann manifold and employ a chordal distance as the means for comparing subspaces. The matching process is accelerated using a coarse to fine strategy. Experiments on the FERET database suggest that the proposed method yields excellent results using both holistic and local features. Specifically, on the FERET Dup2 data set, our proposed method achieves 83.8% rank 1 recognition: to our knowledge the currently the best result among all non-trained methods. Evidence is also presented that peak recognition performance is achieved using roughly 100 distinct perturbed images.