In this paper we introduce a natural image prior that directly represents a Gaussiansmoothed version of the natural image distribution. We include our prior in a formulation of image restoration as a Bayes estimator that also allows us to solve noise-blind image restoration problems. We show that the gradient of our prior corresponds to the mean-shift vector on the natural image distribution. I...

The goal of this project was to develop a fast video image segmentation routine which could be used as a preprocessing step for motion tracking. We chose mean shift [1] as the primary algorithm. Our implementation includes several enhancements including dynamically adjusting the kernel bandwidth based on the overall level of image noise, and keeping a cache of past moves to avoid repeated compu...

Generation of a reference standard from multiple manually annotated datasets is a non-trivial problem. This paper discusses the weighted averaging of 3D open curves, which we used to generate a reference standard for vessel tracking data. We show how weighted averaging can be implemented by applying the Mean Shift algorithm to paths, and discuss the details of our implementation. Our approach c...

The Hough Transform is a well known robust technique to infer shapes from a set of spatial points (Hough (1962); Duda and Hart (1972); Ballard (1981); Goldenshluger and Zeevi (2004); Dattner (2009)). Having a parametric form of the pattern of interest w.r.t. a latent variable Θ, the Hough transform computes an estimate of the density function of Θ using a histogram. Maxima of the histogram are ...

This paper proposes a new variational bound optimization framework for incorporating spatial prior information to the mean shift-based data-driven mode analysis, offering flexible control of the mean shift convergence. Two forms of Gaussian spatial priors are considered. Attractive prior pulls the convergence toward a desired location. Repulsive prior pushes away from such a location. Using a g...

The mean-shift clustering is an efficient technique for color image segmentation by dividing an image into homogeneous regions. The main drawback of mean-shift clustering is to use a fixed scale, which directly determines to use a fixed homogeneity. Since each region could have different homogeneity, using a fixed scale has a problem to segment well. To resolve this problem, we incorporate mult...

The research of its convergence of Mean Shift algorithm is the foundation of its application. Comaniciu and Li Xiang-ru have respectively provided the proof for the convergence of Mean Shift but they both made a mistake in their proofs. In this paper, the imprecise proofs existing in some literatures are firstly pointed out. Then, the local convergence is proved in a new way and the condition o...

The conventional mean shift algorithm has been known to be sensitive to selecting a bandwidth. We present a robust mean shift algorithm with heterogeneous node weights that come from a geometric structure of a given data set. Before running MS procedure, we reconstruct un-normalized weights (a rough surface of data points) from the Delaunay Triangulation. The un-normalized weights help MS to av...

Epanechnikov Mean Shift is a simple yet empirically very effective algorithm for clustering. It localizes the centroids of data clusters via estimating modes of the probability distribution that generates the data points, using the ‘optimal’ Epanechnikov kernel density estimator. However, since the procedure involves non-smooth kernel density functions, the convergence behavior of Epanechnikov ...

Given sets of observations of training and test data, we consider the problem of re-weighting the training data such that its distribution more closely matches that of the test data. We achieve this goal by matching covariate distributions between training and test sets in a high dimensional feature space (specifically, a reproducing kernel Hilbert space). This approach does not require distrib...