Robust Shape Normalization Based on Implicit Representations

Tingting Jiang (1) , Carlo Tomasi (2)

1 LEAR - Learning and recognition in vision
2 Department of Computer Science

We introduce a new shape normalization method based on implicit shape representations. The proposed method is robust with respect to deformations and invariant to similarity transformations (translation, isotropic scaling and rotation). The new method has been tested and compared to the classical shape normalization method and previous work in terms of aligning groups of shapes with deformations.

Keywords: image representation; shape recognition

Domains: Computer Vision and Pattern Recognition [cs.CV]

Tingting Jiang, Carlo Tomasi. Robust Shape Normalization Based on Implicit Representations. ICPR 2008 - IEEE International Conference on Pattern Recognition, Oct 2008, Tampa, United States. pp.1-4, ⟨10.1109/ICPR.2008.4761202⟩. ⟨inria-00548661⟩