Implicit Manipulation of Constraint Sets for Geometric Matching under 2d Translation and Rotation
نویسنده
چکیده
This paper presents a new algorithm in the RAST family of algorithms. RAST algorithms perform geometric matching by exploring intersections between query regions and constraint sets in the space of possible model transformations. RAST algorithms are closely related to hierarchical Hough transformations but have more desirable geometric and combinatorial properties for object recognition applications. Previous applications of the RAST algorithm were limited by the need to represent constraint sets explicitly in transformation space. This paper introduces methods of manipulating the constraint sets arising from image and model point correspondences implicitly, allowing RAST methods to be applied in situations where no simple representation of constraint sets is algebraically possible. The method is demonstrated for exact bounded error matching in the space of 2D translations and rotations.
منابع مشابه
Protein Structure-Structure Alignment with Discrete Fr'echet Distance
Matching two geometric objects in two-dimensional (2D) and three-dimensional (3D) spaces is a central problem in computer vision, pattern recognition, and protein structure prediction. In particular, the problem of aligning two polygonal chains under translation and rotation to minimize their distance has been studied using various distance measures. It is well known that the Hausdorff distance...
متن کاملProbabilistic matching of sets of polygonal curves∗
Analysis and comparison of geometric shapes are of importance in various application areas within computer science, e.g., pattern recognition and computer vision. The general situation in a shape matching problem is that we are given two shapes A and B and a certain class T of allowable transformations and we want to transform B optimally so that the transformed image of B is as close to A as p...
متن کاملNew Algorithms for 2D and 3D Point Matching: Pose Estimation and Correspondence
A fundamental open problem in computer vision-determining pose and correspondence between two sets of points in spaceis solved with a novel, robust and easily implementable algorithm. The technique works on noisy point sets that may be of unequal sizes and may differ by non-rigid transformations. A 2D variation calculates the pose between point sets related by an affine transformation-translati...
متن کاملRange Image Integration Based on 2D Shape Matching
This paper describes an integration technique of multiple range images. The range image integration is to estimate transfer parameters between two shape data, and it consists of an initial guess and a correction of the parameters. The initial guess process is based on 2D shape integration under a condition that the rotation axis of an object is parallel to one of sensor coordinate axes. After t...
متن کاملGeometric Invariant Curve and Surface Normalization
In this work, a geometric invariant curve and surface normalization method is presented. Translation, scale and shear are normalized by Principal Component Analysis (PCA) whitening. Independent Component Analysis (ICA) and the third order moments are then employed for rotation and reflection normalization. By applying this normalization, curves and surfaces that are related by geometric transfo...
متن کامل