We find dense point-to-point correspondences between two surfaces corresponding to different postures of the same articulated object in a fully automatic way. The approach requires no prior knowledge about the shapes being registered. Furthermore, the approach does not require any user-specified parameters. We register possibly incomplete triangular meshes. We model the deformations of an object as isometries and solve the correspondence problem by aligning the intrinsic geometries of the manifolds in a suitable space. We apply the technique to segment the surface into near-rigid components.

Additional Metadata
Persistent URL dx.doi.org/10.1109/ICCVW.2009.5457426
Conference 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops 2009
Citation
Wuhrer, S. (Stefanie), Shu, C. (Chang), & Bose, P. (2009). Posture invariant correspondence of triangular meshes in shape space. Presented at the 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops 2009. doi:10.1109/ICCVW.2009.5457426