Learning Vocabularies over a Fine Quantization

A novel similarity measure for bag-of-words type large scale image retrieval is presented. The similarity function is learned in an unsupervised manner, requires no extra space over the standard bag-of-words method and is more discriminative than both L2-based soft assignment and Hamming embedding. The novel similarity function achieves mean average precision that is superior to any result published in the literature on the standard Oxford 5k, Oxford 105k and Paris datasets/protocols. We study the effect of a fine quantization and very large vocabularies (up to 64 million words) and show that the performance of specific object retrieval increases with the size of the vocabulary. This observation is in contradiction with previously published results. We further demonstrate that the large vocabularies increase the speed of the tf-idf scoring step.     

Repetitive interpolation: A robust algorithm for DTM generation from Aerial Laser Scanner Data in forested terrain

We present a new algorithm for digital terrain model (DTM) generation from an airborne laser scanning point cloud, called repetitive interpolation (REIN). It is especially applicable in steep, forested areas where other filtering algorithms typically have problems distinguishing between ground returns and off-ground points reflected in the vegetation. REIN can produce a DTM either in a vector grid or in a TIN data structure. REIN is applied after an initial filtering, which involves removal of all negative outliers and removal of many, but not necessarily all, off-ground points by some existing filtering algorithm. REIN makes use of the redundancy in the initially filtered point cloud (FPC) in order to mitigate the effect of the residual off-ground points. Multiple independent random samples are taken from the initial FPC. From each sample, ground elevation estimates are interpolated at individual DTM locations. Because the lower bounds of the distributions of the elevation estimates at each DTM location are almost insensitive to positive outliers, the true ground elevations can be approximated by adding the global mean offset to the lower bounds, which is estimated from the data. The random sampling makes REIN unique among the methods of filtering airborne laser data. While other filters behave deterministically, always generating a filter error in special situations, in REIN, because of its random aspects, these errors do not occur in each sample, and typically cancel out in the final computation of DTM elevations. Reduction of processing time by parallelization of REIN is possible. REIN was tested in a test area of 2 hectares, encompassing steep relief covered by mixed forest. An Optech ALTM 1020 lidar was used, with a flying height of 260-300 m above the ground, the beam divergence was 0.3 mrad, and the obtained point cloud density for the last returns was 8.5 m^− 2. A DTM grid was generated with 1 m horizontal resolution. The root mean square elevation error of the DTM ranged between ± 0.16 m and ± 0.37 m, depending on REIN sampling rate and number of samples taken, the lowest value achieved with 4 samples and using a 23% sampling rate. The paper also gives a short overview on existing filtering algorithms.     

Neural network models for group behavior prediction: a case of soccer match attendance

Soccer match attendance is an example of group behavior with noisy context that can only be approximated by a limited set of quantifiable factors. However, match attendance is representative of a wider spectrum of context-based behaviors for which only the aggregate effect of otherwise individual decisions is observable. Modeling of such behaviors is desirable from the perspective of economics, psychology, and other social studies with prospective use in simulators, games, product planning, and advertising. In this paper, we evaluate the efficiency of different neural network architectures as models of context in attendance behavior by comparing the achieved prediction accuracy of a multilayer perceptron (MLP), an Elman recurrent neural network (RNN), a time-lagged feedforward neural network (TLFN), and a radial basis function network (RBFN) against a multiple linear regression model, an autoregressive moving average model with exogenous inputs, and a naive cumulative mean model. We show that the MLP, TLFN, and RNN are superior to the RBFN and achieve comparable prediction accuracy on datasets of three teams from the English Football League Championship, which indicates weak importance of context transition modeled by the TLFN and the RNN. The experiments demonstrate that all neural network models outperform linear predictors by a significant margin. We show that neural models built on individual datasets achieve better performance than a generalized neural model constructed from pooled data. We analyze the input parameter influences extracted from trained networks and show that there is an agreement between nonlinear and linear measures about the most significant attributes.

     

Two-level control strategy of an eight link biped walking model

This paper presents an adaptive two-level control strategy for a biped walking model and demonstrates its performance in a wide range of walking modes with considerably diverse model and control parameter settings. Proposed control strategy inherits a push off that resembles considerably to forceful extension of the trailing leg during push off in human locomotion and represents a very important source of forward propulsion. Extensive simulations have shown that adjustments in the push off related parameter on higher between-step control level after each step enable evolution of various walking modes of the biped walker at selected walking speeds and distinctive gait patterns. It also allows us to investigate the changes in gait kinematics and kinetics of the biped walking model due to changes in gait velocity, torso inclination and propulsion distribution profiles.     

Structural Properties of Recurrent Neural Networks

In this article we research the impact of the adaptive learning process of recurrent neural networks (RNN) on the structural properties of the derived graphs. A trained fully connected RNN can be converted to a graph by defining edges between pairs od nodes having significant weights. We measured structural properties of the derived graphs, such as characteristic path lengths, clustering coefficients and degree distributions. The results imply that a trained RNN has significantly larger clustering coefficient than a random network with a comparable connectivity. Besides, the degree distributions show existence of nodes with a large degree or hubs, typical for scale-free networks. We also show analytically and experimentally that this type of degree distribution has increased entropy.     

Cooperative Coding for OFDM Systems

In this paper, we study Orthogonal Frequency Division Multiplexing (OFDM) systems with cooperative coding over frequency selective Rayleigh fading channels. We derive the pairwise error probability (PEP) for the block-fading OFDM channel model. We use the derived PEP to get an upper bound on the frame error probability for the coded cooperative OFDM system. This bound is then utilized in the study of the diversity and coding gains achievable through cooperative coding in OFDM systems for various inter-user channel qualities. We consider the design of cooperative convolutional codes based on the principle of overlays and provide simulation results for different cooperation scenarios. We observe significant gains over conventional non-cooperative OFDM systems. Finally, based on some simple approximations, we provide guidelines for the choice of partners in coded cooperative OFDM systems.     

Wide view supertwisted nematic liquid-crystal light shutter

We present a novel concept of a wide viewing angle liquid-crystal display (LCD) light shutter, taking advantage of the highly symmetric 180 degrees -twisted liquid-crystal (LC) configuration and the light propagation starting with a single mode. In a specific case of the high voltage driven LCD light shutter, the resulting very high overall symmetry allows for the efficient angular compensation of the light attenuation in the optically closed state using just a simple, specifically tuned, negative-birefringent c-plate retarder. The latter can be adapted to compensate for the positive optical birefringence of the homeotropically aligned LC molecules as well as also for a great deal of the angular dependence of the crossed polarizers. We show that the additional partial compensation of the angular dependence of the crossed polarizers can be explained by a combined effect of the positive birefringence of the homeotropically aligned LC molecules, negative-birefringence c-plate retarder, and an in-plane residual positive birefringence (a-plate) of the boundary layers of the LC molecules.     

Quaternary Ti–20Nb–10Zr–5Ta alloy during immersion in simulated physiological solutions: formation of layers,dissolution and biocompatibility

Samples of the quaternary Ti–20Nb–10Zr–5Ta alloy were immersed in Hanks’ simulated physiological solution and in minimum essential medium (MEM) for 25 days. Samples of Ti metal served as controls. During immersion, the concentration of ions dissolved in MEM was measured by inductively coupled plasma mass spectrometry, while at the end of the experiment the composition of the surface layers was analyzed by X-ray photoelectron spectroscopy, and their morphology by scanning electron microscopy equipped for chemical analysis. The surface layer formed during immersion was comprised primarily of TiO₂ but contained oxides of alloying elements as well. The degree of oxidation differed for different metal cations; while titanium achieved the highest valency, tantalum remained as the metal or is oxidized to its sub-oxides. Calcium phosphate was formed in both solutions, while formation of organic-related species was observed only in MEM. Dissolution of titanium ions was similar for metal and alloy. Among alloying elements, zirconium dissolved in the largest quantity. The long-term effects of alloy implanted in the recipient’s body were investigated in MEM, using two types of human cells—an osteoblast-like cell line and immortalized pulmonary fibroblasts. The in vitro biocompatibility of the quaternary alloy was similar to that of titanium, since no detrimental effects on cell survival, induction of apoptosis, delay of growth, or change in alkaline phosphatase activity were observed on incubation in MEM.     

Planung und Bauausführung des Al-Sadd Stadions in Doha,Katar

Design and construction of the Al-Sadd Stadium in Doha, Qatar. The stadium of Al-Sadd Sports Club, which is the most famous soccer club of the state of Qatar, has recently been upgraded and in large parts been reconstructed from scratch within a period of 14 months. The paper describes some specific features of the design and the construction of this project. The issue of lateral torsional buckling of the curved roof structure and the respective calculations are addressed in detail.