Learn More
Previous empirical studies have shown that information along visual contours is known to be concentrated in regions of high magnitude of curvature, and, for closed contours, segments of negative curvature (i.e., concave segments) carry greater perceptual relevance than corresponding regions of positive curvature (i.e., convex segments). Lately, Feldman and(More)
This article addresses 2-dimensional layout of high-dimensional biomedical datasets, which is useful for browsing them efficiently. We employ the Isomap technique, which is based on classical MDS (multi-dimensional scaling) but seeks to preserve the intrinsic geometry of the data, as captured in the geodesic manifold distances between all pairs of data(More)
Stream compaction is an important parallel computing primitive that produces a reduced (compacted) output stream consisting of only valid elements from an input stream containing both invalid and valid elements. Computing on this compacted stream rather than the mixed input stream leads to improvements in performance, load balancing, and memory footprint.(More)
We present a method for improving robustness in featurebased tracking of human motion. Motion flows of features estimated by a standard tracker are modified to be coherent with neighboring ones. This coherence constraint is computed based on a smooth approximation to initial motion flows computed by the tracker. With this tracking results, we demonstrate(More)
We approach the problem of example-based motion synthesis by transforming motion data into a vector space representation. This allows many techniques successful for stationary object synthesis applicable to that of motion. Especially, by separating generation of motion into a time-consuming preprocess and a fast process, it lets on-the-fly motion synthesis(More)
Somoclu is a massively parallel tool for training self-organizing maps on large data sets written in C++. It builds on OpenMP for multicore execution, and on MPI for distributing the workload across the nodes in a cluster. It is also able to boost training by using CUDA if graphics processing units are available. A sparse kernel is included, which is useful(More)
This paper addresses an issue of solving customers’ problems when applying evolutionary computation. Rather than the seemingly more impressive approach of wow-it-allevolved-from-nothing, tinkering with exisiting models can be a more pragmatic approach in doing so. Using interactive evolution, we experimentally validate this point on setting parameters of a(More)