Understanding patterns of variation from rawmeasurement data remains a central goal of shape analysis. Such an understanding reveals which elements are repeated, or how elements can be derived as structured variations from a common base element. We investigate this problem in the context of 3D acquisitions of buildings. Utilizing a set of template models, we discover geometric similarities acro...

Active Shape Model (ASM) is one of the most popular methods for image alignment. To improve its matching accuracy, in this paper, ASM searching method is combined with a simplified Elastic Bunch Graph Matching (EBGM) algorithm. Considering that EBGM is too timeconsuming, landmarks are grouped into contour points and inner points, and inner points are further separated into several groups accord...

The quality of 3D object transformation very often relies on the quality of the matching between the two objects. Geometric considerations regarding scale, rotation and translation, and user defined constraints are used to control the matching. Nevertheless, more complex measures regarding volumetric or topological similarity between the shapes are more difficult to define even manually. Choosi...

Shape matching or recognition is computation intensity work in shape analysis. Paper [3] proposed an efficient shape matching method using shape contexts (SC). In this project, wavelet transform is used to scale or compress images into much smaller sizes and then shape context algorithm is used to match shapes. Simulation results show that the matching speed is faster and matching results are m...

This paper describes a Bayesian graph matching algorithm for data-mining from large structural databases. The matching algorithm uses edge-consistency and node attribute similarity to determine the a posteriori probability of a query graph for each of the candidate matches in the database. The node feature-vectors are constructed by computing normalised histograms of pairwise geometric attribut...

Many visual matching algorithms can be described in terms of the features and the inter-feature distance or metric. The most commonly used metric is the sum of squared differences (SSD), which is valid from a maximum likelihood perspective when the real noise distribution is Gaussian. However, we have found experimentally that the Gaussian noise distribution assumption is often invalid. This im...