The (k-)nearest neighbor searching has very high computational costs. The algorithms presented for nearest neighbor search in high dimensional spaces have have suffered from curse of dimensionality, which affects either runtime or storage requirements of the algorithms terribly. Parallelization of nearest neighbor search is a suitable solution for decreasing the workload caused by nearest neigh...

This paper presents a neighbor-search technique to be used in a GPU-based particle simulation framework. Neighbor searching is usually the most computationally expensive step in particle simulations. The usual approach is to subdivide the simulation space to speed up neighbor search. Here, we present a grid-based neighbor-search technique designed to work on programmable graphics hardware.

The robustness (sensitivity to violation of assumptions) of the maximum-likelihood and neighbor-joining methods was examined using simulation. Maximum likelihood and neighbor joining were implemented with Jukes-Cantor, Kimura, and gamma models of DNA substitution. Simulations were performed in which the assumptions of the methods were violated to varying degrees on three model four-taxon trees....

The neighbor-exclusion principle is one of the most general and interesting rules describing intercalative DNA binding by small molecules. It suggests that such binding can only occur at every other base-pair site, reflecting a very large negative cooperativity in the binding process. We have carried out molecular mechanics and molecular dynamics simulations to study intercalation complexes bet...

The nearest-neighbor based document skew detection methods do not require the presence of a predominant text area, and are not subject to skew angle limitation. However, the accuracy of these methods is not perfect in general. In this paper, we present an improved nearest-neighbor based approach to perform accurate document skew estimation. Size restriction is introduced to the detection of nea...

OF THE THESIS Aren’t we all nearest neighbors?

Matching persons across non-overlapping cameras is a rather challenging task. Thus, successful methods often build on complex feature representations or sophisticated learners. A recent trend to tackle this problem is to use metric learning to find a suitable space for matching samples from different cameras. However, most of these approaches ignore the transition from one camera to the other. ...