Recognition of 3D objects from different viewpoints is a difficult problem. In this paper, we propose a new method to recognize 3D range images by matching local surface descriptors. The input 3D surfaces are first converted into a set of local shape descriptors computed on surface patches defined by detected salient features. We compute the similarities between input 3D images by matching their descriptors with a pyramid kernel function. The similarity matrix of the images is used to train for classification using SVM, and new images can be recognized by comparing with the training set. The approach is evaluated on both synthetic and real 3D data with complex shapes.

This paper introduces a novel kernel-based method for template tracking in video sequences. The method is derived for a general warping transformation, and its application to affine motion tracking is further explored. Our approach is based on maximization of the multi-kernel Bhattacharyya coefficient with respect to the warp parameters. We explicitly compute the gradient of the similarity functional, and use a quasi-Newton procedure for optimization. Additionally, we consider a simple extension of the method that employs an illumination model correction to allow tracking under varying lighting conditions. The resulting tracking procedure is evaluated on a number of examples including large templates tracking non-rigidly moving textured areas.

Surface matching is a common task in computer graphics and computer vision. In this paper, we introduce a novel algorithm that aligns scanned point-based sufaces to the related 3 0 model without assumptions about their relative position and orientation. Our method estimates the transformation that matches two surfaces from a pair of corresponding points using their suface positions, nonnals and principal curvature directions. This approximate alignment is used as the input of ZCP algorithm to rejne the registration of the suflace to the model. The algorithm is easy to implement and proved to be reliable even with coarse point clouds. Furthermore, with the help of multi-scale feature points, we can improve the reliability of our surface matching algorithm.

We propose an out-of-core method for creating semi-regular surface representations from large input surface meshes. Our approach is based on a streaming implementation of the MAPS remesher of Lee et al. [18]. Our remeshing procedure consists of two stages. First, a simplification process is used to obtain the base domain. During simplification, we maintain the mapping information between the input and the simplified meshes. The second stage of remeshing uses the mapping information to produce samples of the output semi-regular mesh. The out-of-core operation of our method is enabled by the synchronous streaming of a simplified mesh and the mapping information stored at the original vertices. The synchronicity of two streaming buffers is maintained using a specially designed write strategy for each buffer. Experimental results demonstrate the remeshing performance of the proposed method, as well as other applications that use the created mapping between the simplified and the original surface representations.

A hybrid mesh is a multiresolution surface representation that combines advantages from regular and irregular meshes. Irregular operations allow a hybrid mesh to change topology throughout the hierarchy and approximate detailed features at multiple scales. A preponderance of regular refinements allows for efficient data-structures and processing algorithms. We provide a user driven procedure for creating a hybrid mesh from scanned geometry and present a progressive hybrid mesh compression algorithm.

Normal meshes are new fundamental surface descriptions inspired by differential geometry. A normal mesh is a multiresolution mesh where each level can be written as a normal offset from a coarser version. Hence the mesh can be stored with a single float per vertex. We present an algorithm to approximate any surface arbitrarily closely with a normal semi-regular mesh. Normal meshes can be useful in numerous applications such as compression, filtering, rendering, texturing, and modeling.

We generalize basic signal processing tools such as downsampling, upsampling, and filters to irregular connectivity triangle meshes.This is accomplished through the design of a non-uniform relaxation procedure whose weights depend on the geometry and we show its superiority over existing schemes whose weights depend only on connectivity.This is combined with known mesh simplification methods to build subdivision and pyramid algorithms.We demonstrate the power of these algorithms through a number of application examples including smoothing, enhancement, editing, and texture mapping.