A Multi-View Nonlinear Active Shape Model Using Kernel PCA

• DOI: 10.5244/c.13.48
• Publication Date: 2013-02-05T17:17:44Z
• Authors: S. Romdhani; S. Gong; A. Psarrou
• Keywords: Kernel (algebra); Heat kernel signature; Feature (linguistics)

Recovering the shape of any 3D object using multiple 2D views requires establishing correspondence between feature points at different views. However changes in viewpoint introduce self-occlusions, resulting nonlinear variations and inconsistent features Here we a multi-view model utilising view-dependent constraint without explicit reference to structures. For transformation, adopt Kernel PCA based on Support Vector Machines.