Efficient tracking of human poses using a manifold hierarchy

In this paper a 3D human pose tracking framework is presented. A new dimensionality reduction method (Hierarchical Temporal Laplacian Eigenmaps) is introduced to represent activities in hierarchies of low dimensional spaces. Such a hierarchy provides increasing independence between limbs, allowing higher flexibility and adaptability that result in improved accuracy. Moreover, a novel deterministic optimisation method (Hierarchical Manifold Search) is applied to estimate efficiently the position of the corresponding body parts. Finally, evaluation on public datasets such as HumanEva demonstrates that our approach achieves a 62.5mm-65mm average joint error for the walking activity and outperforms state-of-the-art methods in terms of accuracy and computational cost.

Vehicle-barrier tracking of a scaled crash test for roadside barrier design

In this paper the tracking system used to perform a scaled vehicle-barrier crash test is reported. The scaled crash test was performed as part of a wider project aimed at designing a new safety barrier making use of natural building materials. The scaled crash test was designed and performed as a proof of concept of the new mass-based safety barriers and the study was composed of two parts: the scaling technique and of a series of performed scaled crash tests. The scaling method was used for 1) setting the scaled test impact velocity so that energy dissipation and momentum transferring, from the car to the barrier, can be reproduced and 2) predicting the acceleration, velocity and displacement values occurring in the full-scale impact from the results obtained in a scaled test. To achieve this goal the vehicle and barrier displacements were to be recorded together with the vehicle accelerations and angular velocities. These quantities were measured during the tests using acceleration sensors and a tracking system. The tracking system was composed of a high speed camera and a set of targets to measure the vehicle linear and angular velocities. A code was developed to extract the target velocities from the videos and the velocities obtained were then compared with those obtained integrating the accelerations provided by the sensors to check the reliability of the method.

Robust visual tracking using structurally random projection and weighted least squares

Sparse representation based visual tracking approaches have attracted increasing interests in the community in recent years. The main idea is to linearly represent each target candidate using a set of target and trivial templates while imposing a sparsity constraint onto the representation coefficients. After we obtain the coefficients using L1-norm minimization methods, the candidate with the lowest error, when it is reconstructed using only the target templates and the associated coefficients, is considered as the tracking result. In spite of promising system performance widely reported, it is unclear if the performance of these trackers can be maximised. In addition, computational complexity caused by the dimensionality of the feature space limits these algorithms in real-time applications. In this paper, we propose a real-time visual tracking method based on structurally random projection and weighted least squares techniques. In particular, to enhance the discriminative capability of the tracker, we introduce background templates to the linear representation framework. To handle appearance variations over time, we relax the sparsity constraint using a weighed least squares (WLS) method to obtain the representation coefficients. To further reduce the computational complexity, structurally random projection is used to reduce the dimensionality of the feature space while preserving the pairwise distances between the data points in the feature space. Experimental results show that the proposed approach outperforms several state-of-the-art tracking methods.

Tracking Change in Northern Ireland Policing: Temporal Phases and Key Themes

Purpose
– The purpose of this paper is to explore and explain the change process in Northern Ireland policing through an analysis of temporally bracketed change phases and key change delivery themes ranging from 1996 to 2012.

Design/methodology/approach
– The research approach adopted is process based, longitudinal and multi-method, utilising “temporal bracketing” to determine phases of change and conjunctural reasoning to unravel the systematic factors interacting over time, within the case.

Findings
– The paper identifies and temporally brackets four phases of change: “Tipping point”; “Implementation, Symbolic Modification and Resistance”; “Power Assisted Steering”; and “A Return to Turbulence”, identifies four themes that emerge from RUC-PSNI experience: the role of adaptive leadership; pace and sequencing of change implementation; sufficient resourcing; and the impact of external agents acting as boundary spanners, and comments on the prominence of these themes through the phases. The paper goes on to reflect upon how these phases and themes inform our understanding of organisational change within policing organisations generally and within politically pressurised transition processes.

Originality/value
– The contribution of the paper lies in the documentation of an almost unique organisational case in an environmentally forced change process. In this it contains lessons for other organisations facing similar, if less extreme challenges and presents an example of intense change analysed longitudinally.

Similarity between Positronium-Atom and Electron-Atom Scattering

We employ the impulse approximation for a description of positronium-atom scattering. Our analysis and calculations of Ps-Kr and Ps-Ar collisions provide a theoretical explanation of the similarity between the cross sections for positronium scattering and electron scattering for a range of atomic and molecular targets observed by S. J. Brawley et al. [Science 330, 789 (2010)].

Positronium-atom scattering at low energies

A pseudopotential for positronium-atom interaction, based on electron-atom and positron-atom phase shifts, is constructed, and the phase shifts for Ps-Kr and Ps-Ar scattering are calculated. This approach allows us to extend the Ps-atom cross sections, obtained previously in the impulse approximation [I. I. Fabrikant and G. F. Gribakin, Phys. Rev. Lett. 112, 243201 (2014)], to energies below the Ps ionization threshold. Although experimental data are not available in this low-energy region, our results describe well the tendency of the measured cross sections to drop with decreasing velocity at v < 1 a.u. Our results show that the effect of the Ps-atom van der Waals interaction is weak compared to the polarization interaction in electron-atom and positron-atom scattering. As a result, the Ps scattering length for both Ar and Kr is positive, and the Ramsauer-Townsend minimum is not observed for Ps scattering from these targets. This makes Ps scattering quite different from electron scattering in the low-energy region, in contrast to the intermediate energy range from the Ps ionization threshold up to v ∼ 2 a.u., where the two are similar.

Density functional study of the C atom adsorption on the α-Fe 2O3 (001) surface

The adsorption of C atoms on the α-Fe₂O₃ (001) surface was studied based on density function theory (DFT), in which the exchange-correlation potential was chosen as the PBE (Perdew, Burke and Ernzerhof) generalized gradient approximation (GGA) with a plane wave basis set. Upon the optimization on different adsorption sites with coverage of 1/20 and 1/5 ML, it was found that the adsorption of C atoms on the α-Fe ₂O₃ (001) surface was chemical adsorption. The coverage can affect the adsorption behavior greatly. Under low coverage, the most stable adsorption geometry lied on the bridged site with the adsorption energy of about 3.22 eV; however, under high coverage, it located at the top site with the energy change of 8.79 eV. Strong chemical reaction has occurred between the C and O atoms at this site. The density of states and population analysis showed that the s, p orbitals of C and p orbital of O give the most contribution to the adsorption bonding. During the adsorption process, O atom shares the electrons with C, and C can only affect the outermost and subsurface layers of α-Fe₂O₃; the third layer can not be affected obviously. Copyright © 2008 Chinese Journal of Structural Chemistry.

Tracking smart grid hackers

The next-generation smart grid will rely highly on telecommunications infrastructure for data transfer between various systems. Anywhere we have data transfer in a system is a potential security threat. When we consider the possibility of smart grid data being at the heart of our critical systems infrastructure it is imperative that we do all we can to ensure the confidentiality, availability and integrity of the data. A discussion on security itself is outside the scope of this paper, but if we assume the network to be as secure as possible we must consider what we can do to detect when that security fails, or when the attacks comes from the inside of the network. One way to do this is to setup a hacker-trap, or honeypot. A honeypot is a device or service on a network which appears legitimate, but is in-fact a trap setup to catch breech attempts. This paper identifies the different types of honeypot and describes where each may be used. The authors have setup a test honeypot system which has been live for some time. The test system has been setup to emulate a device on a utility network. The system has had many hits, which are described in detail by the authors. Finally, the authors discuss how larger-scale systems in utilities may benefit from honeypot placement.

More nonlocality with less entanglement in a tripartite atom-optomechanical system

Quantum effects in hybrid atomic optomechanics in a system comprising a cloud of atoms and a mobile mirror mediated by a single-mode cavity are studied. Tripartite non-locality is observed in the atom-light-mirror system, as demonstrated by the violation of the Mermin-Klyshko (MK) inequality. It has been shown [C. Genes, et al., PRA 77, 050307 (R) (2008)] that tripartite entanglement is optimized when the cavity is resonant with the anti-Stokes sideband of the driving laser and the atomic frequency matches the Stokes one. However, we show that this is not the case for the nonlocality. The MK function achieves minima when the atoms are resonant with both the Stokes and anti-Stokes sidebands, and unexpectedly, we find violation of the MK inequality only in a parameter region where entanglement is far from being maximum. A negative relation exists between nonlocality and entanglement with consideration of the possibility of bipartite nonlocality in the violation of the MK inequality. We also study the non-classicality of the mirror by post-selected measurements, e.g. Geiger-like detection, on the cavity and/or the atoms. We show that with feasible parameters Geiger-like detection on the atoms can effectively induce mechanical non-classicality.

Visual attention to food cues in obesity: An eye‐tracking study.

Objective
Based on the theory of incentive sensitization, the aim of this study was to investigate differences in attentional processing of food-related visual cues between normal-weight and overweight/obese males and females.

Methods
Twenty-six normal-weight (14M, 12F) and 26 overweight/obese (14M, 12F) adults completed a visual probe task and an eye-tracking paradigm. Reaction times and eye movements to food and control images were collected during both a fasted and fed condition in a counterbalanced design.

Results
Participants had greater visual attention towards high-energy-density food images compared to low-energy-density food images regardless of hunger condition. This was most pronounced in overweight/obese males who had significantly greater maintained attention towards high-energy-density food images when compared with their normal-weight counterparts however no between weight group differences were observed for female participants.

Conclusions
High-energy-density food images appear to capture visual attention more readily than low-energy-density food images. Results also suggest the possibility of an altered visual food cue-associated reward system in overweight/obese males. Attentional processing of food cues may play a role in eating behaviors thus should be taken into consideration as part of an integrated approach to curbing obesity.