Function-based single and dual point haptic interaction in Cyberworlds

Polygon and point based models dominate virtual reality. These models also affect haptic rendering algorithms, which are often based on collision with polygons. With application to dual point haptic devices for operations like grasping, complex polygon and point based models will make the collision detection procedure slow. This results in the system not able to achieve interactivity for force rendering. To solve this issue, we use mathematical functions to define and implement geometry (curves, surfaces and solid objects), visual appearance (3D colours and geometric textures) and various tangible physical properties (elasticity, friction, viscosity, and force fields). The function definitions are given as analytical formulas (explicit, implicit and parametric), function scripts and procedures. We proposed an algorithm for haptic rendering of virtual scenes including mutually penetrating objects with different sizes and arbitrary location of the observer without a prior knowledge of the scene to be rendered. The algorithm is based on casting multiple haptic rendering rays from the Haptic Interaction Point (HIP), and it builds a stack to keep track on all colliding objects with the HIP. The algorithm uses collision detection based on implicit function representation of the object surfaces. The proposed approach allows us to be flexible when choosing the actual rendering platform, while it can also be easily adopted for dual point haptic collision detection as well as force and torque rendering. The function-defined objects and parts constituting them can be used together with other common definitions of virtual objects such as polygon meshes, point sets, voxel volumes, etc. We implemented an extension of X3D and VRML as well as several standalone application examples to validate the proposed methodology. Experiments show that our concern about fast, accurate rendering as well as compact representation could be fulfilled in various application scenarios and on both single and dual point haptic devices.

An adaptive algorithm for finding frequent sets in landmark windows

We consider a CPU constrained environment for finding approximation of frequent sets in data streams using the landmark window. Our algorithm can detect overload situations, i.e., breaching the CPU capacity, and sheds data in the stream to “keep up”. This is done within a controlled error threshold by exploiting the Chernoff-bound. Empirical evaluation of the algorithm confirms the feasibility.

Compact stacked planar inverted-F antenna for passive deep brain stimulation implants

A compact meandered three-layer stacked circular planar inverted-F antenna is designed and simulated at the UHF band (902.75 – 927.25 MHz) for passive deep brain stimulation implants. The UHF band is used because it offers small antenna size, and high data rate. The top and middle radiating layers are meandered, and low cost substrate and superstrate materials are used to limit the radius and height of the antenna to 5 mm and 1.64 mm, respectively. A dielectric substrate of FR-4 of εr= 4.7 and δ= 0.018, and a biocompatible superstrate of silicone of er= 3.7 and d= 0.003 with thickness of 0.2 mm are used in the design. The resonance frequency of the proposed antenna is 918 MHz with a bandwidth of 24 MHz at return loss of −10 dB in free space. The antenna parameter such as 3D gain pattern of the designed antenna within a skin-tissue model is evaluated by using the finite element method. The compactness, wide bandwidth, round shape, and stable characteristics in skin make this antenna suitable for DBS. The feasibility of the wireless power transmission to the implant in the human head is also examined.

Learning from ordered sets and applications in collaborative ranking

Ranking over sets arise when users choose between groups of items. For example, a group may be of those movies deemed 5 stars to them, or a customized tour package. It turns out, to model this data type properly, we need to investigate the general combinatorics problem of partitioning a set and ordering the subsets. Here we construct a probabilistic log-linear model over a set of ordered subsets. Inference in this combinatorial space is highly challenging: The space size approaches (N!/2)6.93145N+1 as N approaches infinity. We propose a split-and-merge Metropolis-Hastings procedure that can explore the state-space efficiently. For discovering hidden aspects in the data, we enrich the model with latent binary variables so that the posteriors can be efficiently evaluated. Finally, we evaluate the proposed model on large-scale collaborative filtering tasks and demonstrate that it is competitive against state-of-the-art methods.

A new dempster-shafer theory-based method with fuzzy targets for fuzzy sets ranking

In this paper, a new Fuzzy Set (FS) ranking method (for type-1 and interval type-2 FSs), which is based on the Dempster-Shafer Theory (DST) of evidence with fuzzy targets, is investigated. Fuzzy targets are adopted to reflect human viewpoints on fuzzy ranking. Two important measures in DST, i.e., the belief and plausibility measures, are used to rank FSs. The proposed approach is evaluated with several benchmark examples. The use of the belief and plausibility measures in fuzzy ranking are discussed and compared. We further analyze the capability of the proposed approach in fulfilling six reasonable fuzzy ordering properties as discussed in [9]-[11].

Recognition from appearance subspaces across image sets of variable scale

Linear subspace representations of appearance variation are pervasive in computer vision. In this paper we address the problem of robustly matching them (computing the similarity between them) when they correspond to sets of images of different (possibly greatly so) scales. We show that the naïve solution of projecting the low-scale subspace into the high-scale image space is inadequate, especially at large scale discrepancies. A successful approach is proposed instead. It consists of (i) an interpolated projection of the low-scale subspace into the high-scale space, which is followed by (ii) a rotation of this initial estimate within the bounds of the imposed “downsampling constraint”. The optimal rotation is found in the closed-form which best aligns the high-scale reconstruction of the low-scale subspace with the reference it is compared to. The proposed method is evaluated on the problem of matching sets of face appearances under varying illumination. In comparison to the naïve matching, our algorithm is shown to greatly increase the separation of between-class and within-class similarities, as well as produce far more meaningful modes of common appearance on which the match score is based.

Face recognition with image sets using manifold density divergence

In many automatic face recognition applications, a set of a person's face images is available rather than a single image. In this paper, we describe a novel method for face recognition using image sets. We propose a flexible, semi-parametric model for learning probability densities confined to highly non-linear but intrinsically low-dimensional manifolds. The model leads to a statistical formulation of the recognition problem in terms of minimizing the divergence between densities estimated on these manifolds. The proposed method is evaluated on a large data set, acquired in realistic imaging conditions with severe illumination variation. Our algorithm is shown to match the best and outperform other state-of-the-art algorithms in the literature, achieving 94% recognition rate on average.

Learning over sets using boosted manifold principal angles (BoMPA)

In this paper we address the problem of classifying vector sets. We motivate and introduce a novel method based on comparisons between corresponding vector subspaces. In particular, there are two main areas of novelty: (i) we extend the concept of principal angles between linear subspaces to manifolds with arbitrary nonlinearities; (ii) it is demonstrated how boosting can be used for application-optimal principal angle fusion. The strengths of the proposed method are empirically demonstrated on the task of automatic face recognition (AFR), in which it is shown to outperform state-of-the-art methods in the literature.

Learning aggregation weights from 3-tuple comparison sets

An important task in multiple-criteria decision making is how to learn the weights and parameters of an aggregation function from empirical data. We consider this in the context of quantifying ecological diversity, where such data is to be obtained as a set of pairwise comparisons specifying that one community should be considered more diverse than another. A problem that arises is how to collect a sufficient amount of data for reliable model determination without overloading individuals with the number of comparisons they need to make. After providing an algorithm for determining criteria weights and an overall ranking from such information, we then investigate the improvement in accuracy if ranked 3-tuples are supplied instead of pairs. We found that aggregation models could be determined accurately from significantly fewer 3-tuple comparisons than pairs. © 2013 IEEE.