Modeling Deformable Surfaces from Single Videos

• To avoid having to explicitly force the coefficients of overlapping patches to be consistent, we express them as which arises from the orthonormality of the modes.• Regularization is achieved by penalizing the coefficients associated to high energy modes, which is done by minimizing where contains the eigenvalues of the training data covariance matrix.

[1]  Vincent Lepetit,et al.  BRIEF: Binary Robust Independent Elementary Features , 2010, ECCV.

[2]  Pascal Fua,et al.  Linear Local Models for Monocular Reconstruction of Deformable Surfaces , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Vincent Lepetit,et al.  Fast Non-Rigid Surface Detection, Registration and Realistic Augmentation , 2008, International Journal of Computer Vision.

[4]  Vincent Lepetit,et al.  Fast Keypoint Recognition Using Random Ferns , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Pascal Fua,et al.  Surface Deformation Models for Nonrigid 3D Shape Recovery , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.