
For FullText PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.

Construction of Appearance Manifold with Embedded ViewDependent Covariance Matrix for 3D Object Recognition
Lina Tomokazu TAKAHASHI Ichiro IDE Hiroshi MURASE
Publication
IEICE TRANSACTIONS on Information and Systems
Vol.E91D
No.4
pp.10911100 Publication Date: 2008/04/01
Online ISSN: 17451361
DOI: 10.1093/ietisy/e91d.4.1091
Print ISSN: 09168532 Type of Manuscript: PAPER Category: Pattern Recognition Keyword: 3D object recognition, appearance manifold, viewdependent covariance matrix, eigenvector interpolation, eigenvalue interpolation, eigenspace,
Full Text: PDF>>
Summary:
We propose the construction of an appearance manifold with embedded viewdependent covariance matrix to recognize 3D objects which are influenced by geometric distortions and quality degradation effects. The appearance manifold is used to capture the pose variability, while the covariance matrix is used to learn the distribution of samples for gaining noiseinvariance. However, since the appearance of an object in the captured image is different for every different pose, the covariance matrix value is also different for every pose position. Therefore, it is important to embed viewdependent covariance matrices in the manifold of an object. We propose two models of constructing an appearance manifold with viewdependent covariance matrix, called the Viewdependent Covariance matrix by trainingPoint Interpolation (VCPI) and Viewdependent Covariance matrix by Eigenvector Interpolation (VCEI) methods. Here, the embedded viewdependent covariance matrix of the VCPI method is obtained by interpolating every trainingpoints from one pose to other trainingpoints in a consecutive pose. Meanwhile, in the VCEI method, the embedded viewdependent covariance matrix is obtained by interpolating only the eigenvectors and eigenvalues without considering the correspondences of each training image. As it embeds the covariance matrix in manifold, our viewdependent covariance matrix methods are robust to any pose changes and are also noise invariant. Our main goal is to construct a robust and efficient manifold with embedded viewdependent covariance matrix for recognizing objects from images which are influenced with various degradation effects.

