Deformable spatio-temporal shape models: extending active shape models to 2D+time

摘要

This paper extends 2D active shape models to 2D+time by presenting a method for modeling and segmenting spatio-temporal shapes (ST-shapes). The modeling part consists of constructing a statistical model of ST-shape parameters. This model describes the principal modes of variation of the ST-shape in addition to constraints on the allowed variations. An active approach is used in segmentation where an initial ST-shape is deformed to better fit the data and the optimal proposed deformation is calculated using dynamic programming. Segmentation results on both synthetic and real data are presented.