Shape Database Classification using Advanced Machine Vision Methods and Stationary Transformed Features

Abstract

The goal of this research is to model an “activity” performed by a group of moving and interacting objects and use these
models for shape detection, tracking and segmentation, finally detecting the shapes. Previous approaches to modeling group
activity include co-occurrence statistics and Dynamic Bayesian Networks, neither of which is applicable when the number of
interacting objects is large. We treat the objects as point objects and propose to model their changing configuration as a moving
and deforming “shape” using ideas from Kendall’s shape theory for discrete landmarks. A continuous state stationary
transformed feature is defined for landmark shape dynamics in an “activity”. The configuration of landmarks at a given time
forms the observation vector and the corresponding shape and scaled Euclidean motion parameters form the hidden state vector.
The simulated results show that the presented method is superior to traditional techniques.