Application of the cross entropy method to the GLVQ algorithm

This paper discusses an alternative approach to parameter optimization of well-known prototype-based learning algorithms (minimizing an objective function via gradient search). The proposed approach considers a stochastic optimization called the cross entropy method (CE method). The CE method is used to tackle efficiently the initialization sensitiveness problem associated with the original generalized learning vector quantization (GLVQ) algorithm and its variants. Results presented in this paper indicate that the CE method can be successfully applied to this kind of problem on real-world data sets. As far as known by the authors, it is the first use of the CE method in prototype-based learning.     

Accelerating Ray Tracing using Constrained Tetrahedralizations

In this paper we introduce the constrained tetrahedralization as a new acceleration structure for ray tracing. A constrained tetrahedralization of a scene is a tetrahedralization that respects the faces of the scene geometry. The closest intersection of a ray with a scene is found by traversing this tetrahedralization along the ray, one tetrahedron at a time. We show that constrained tetrahedralizations are a viable alternative to current acceleration structures, and that they have a number of unique properties that set them apart from other acceleration structures: constrained tetrahedralizations are not hierarchical yet adaptive; the complexity of traversing them is a function of local geometric complexity rather than global geometric complexity; constrained tetrahedralizations support deforming geometry without any effort; and they have the potential to unify several data structures currently used in global illumination.     

Tensor Clustering for Rendering Many‐Light Animations

Rendering animations of scenes with deformable objects, camera motion, and complex illumination, including indirect lighting and arbitrary shading, is a long‐standing challenge. Prior work has shown that complex lighting can be accurately approximated by a large collection of point lights. In this formulation, rendering of animation sequences becomes the problem of efficiently shading many surface samples from many lights across several frames. This paper presents a tensor formulation of the animated many‐light problem, where each element of the tensor expresses the contribution of one light to one pixel in one frame. We sparsely sample rows and columns of the tensor, and introduce a clustering algorithm to select a small number of representative lights to efficiently approximate the animation. Our algorithm achieves efficiency by reusing representatives across frames, while minimizing temporal flicker. We demonstrate our algorithm in a variety of scenes that include deformable objects, complex illumination and arbitrary shading and show that a surprisingly small number of representative lights is sufficient for high quality rendering. We believe out algorithm will find practical use in applications that require fast previews of complex animation.     

Multicamera people tracking with a probabilistic occupancy map

Given two to four synchronized video streams taken at eye level and from different angles, we show that we can effectively combine a generative model with dynamic programming to accurately follow up to six individuals across thousands of frames in spite of significant occlusions and lighting changes. In addition, we also derive metrically accurate trajectories for each one of them.Our contribution is twofold. First, we demonstrate that our generative model can effectively handle occlusions in each time frame independently, even when the only data available comes from the output of a simple background subtraction algorithm and when the number of individuals is unknown a priori. Second, we show that multi-person tracking can be reliably achieved by processing individual trajectories separately over long sequences, provided that a reasonable heuristic is used to rank these individuals and avoid confusing them with one another.     

Graph rigidity, cyclic belief propagation, and point pattern matching

A recent paper \cite{CaeCaeSchBar06} proposed a provably optimal, polynomial time method for performing near-isometric point pattern matching by means of exact probabilistic inference in a chordal graphical model. Its fundamental result is that the chordal graph in question is shown to be \emph{globally rigid}, implying that exact inference provides the \emph{same} matching solution as exact inference in a complete graphical model. This implies that the algorithm is optimal when there is no noise in the point patterns. In this paper, we present a new graph which is also globally rigid but has an advantage over the graph proposed in \cite{CaeCaeSchBar06}: its maximal clique size is smaller, rendering inference significantly more efficient. However, this graph is not chordal and thus standard Junction Tree algorithms cannot be directly applied. Nevertheless, we show that loopy belief propagation in such a graph converges to the optimal solution. This allows us to retain the optimality guarantee in the noiseless case, while substantially reducing both memory requirements and processing time. Our experimental results show that the accuracy of the proposed solution is indistinguishable from that of \cite{CaeCaeSchBar06} when there is noise in the point patterns.     

Biofidelic child head FE model to simulate real world trauma

Biomechanics of human head has been widely studied since several decades. At a mechanical level, the use of engineering allowed investigating injury mechanisms developing numerical models of adult head. For children, the problem is more difficult and evaluating child injury mechanisms using data obtained from scaling adult injury criteria does not account for differences in morphology and structure between adults and children. During growth, child head undergoes different modifications in morphology and structure. The present paper compares the anthropometry and numerical simulations of a child head model based on medical CT scans to a child head model developed by scaling an adult head model using the method proposed by Mertz [H.J. Mertz, A procedure for normalizing impact response data, SAE paper 840884, 1984]. These analysis point out significant differences showing that scaling down an adult head to obtain a child head does not appear relevant. Biofidelic and specific child geometry is needed to investigate child injury mechanisms.     

Detecting reliable gene interactions by a hierarchy of Bayesian network classifiers

The main purpose of a gene interaction network is to map the relationships of the genes that are out of sight when a genomic study is tackled. DNA microarrays allow the measure of gene expression of thousands of genes at the same time. These data constitute the numeric seed for the induction of the gene networks. In this paper, we propose a new approach to build gene networks by means of Bayesian classifiers, variable selection and bootstrap resampling. The interactions induced by the Bayesian classifiers are based both on the expression levels and on the phenotype information of the supervised variable. Feature selection and bootstrap resampling add reliability and robustness to the overall process removing the false positive findings. The consensus among all the induced models produces a hierarchy of dependences and, thus, of variables. Biologists can define the depth level of the model hierarchy so the set of interactions and genes involved can vary from a sparse to a dense set. Experimental results show how these networks perform well on classification tasks. The biological validation matches previous biological findings and opens new hypothesis for future studies.     

Inferring (Biological) Signal Transduction Networks via Transitive Reductions of Directed Graphs

In this paper we consider the p-ary transitive reduction (TR _p ) problem where p>0 is an integer; for p=2 this problem arises in inferring a sparsest possible (biological) signal transduction network consistent with a set of experimental observations with a goal to minimize false positive inferences even if risking false negatives. Special cases of TR _p have been investigated before in different contexts; the best previous results are as follows:

Towards interactive simulation in automotive design

One of the important tasks in Mechanical Engineering is to increase the safety of the vehicle and decrease its production costs. This task is typically solved by means of Multiobjective Optimization, which formulates the problem as a mapping from the space of design variables to the space of target criteria and tries to find an optimal region in these multidimensional spaces. Due to high computational costs of numerical simulations, the sampling of this mapping is usually very sparse and scattered. Combining design of experiments methods, metamodeling, new interpolation schemes and innovative graphics methods, we enable the user to interact with simulation parameters, optimization criteria, and come to a new interpolated crash result within seconds. We denote this approach as Simulated Reality, a new concept for the interplay between simulation, optimization and interactive visualization. In this paper we show the application of Simulated Reality for solution of real life car design optimization problems.

Real-time crowd motion planning

Real-time crowd motion planning requires fast, realistic methods for path planning as well as obstacle avoidance. In a previous work (Morini et al. in Cyberworlds International Conference, pp. 144–151, 2007), we introduced a hybrid architecture to handle real-time motion planning of thousands of pedestrians. In this article, we present an extended version of our architecture, introducing two new features: an improved short-term collision avoidance algorithm, and simple efficient group behavior for crowds. Our approach allows the use of several motion planning algorithms of different precision for regions of varied interest. Pedestrian motion continuity is ensured when switching between such algorithms. To assess our architecture, several performance tests have been conducted, as well as a subjective test demonstrating the impact of using groups. Our results show that the architecture can plan motion in real time for several thousands of characters.