Deformable Spatio-Temporal Shape Modeling
Ghassan Hamarneh · 1999
This Thesis presents a method for modeling spatio-temporal shapes from a training set and demonstrates its application to detecting and segmenting similar shapes in digital images. The modeling part consists of constructing a statistical model of shape parameters. The model obtained describes the principal modes of variation of the spatio-temporal shape in addition to certain constraints on the allowed variations. An active approach is used in segmentation; an initial spatio-temporal shape is deformed to better fit the data and the optimal proposed deformation is calculated using dynamic programming. The results presented show that the proposed method, belonging to the class of Active Shape Models, can successfully detect complex spatio-temporal shapes in noisy data. The Thesis also presents an overview of shape representations and deformable shape models including recent work on Active Shape Models. Keywords: spatio-temporal shapes, spatio-temporal segmentation, shape variation, shap...