Exploring Transfer Learning Approaches for Head Pose Classification from Multi-view Surveillance Images

Head pose classification from surveillance images acquired with distant, large field-of-view cameras is difficult as faces are captured at low-resolution and have a blurred appearance. Domain adaptation approaches are useful for transferring knowledge from the training (source) to the test (target) data when they have different attributes, minimizing target data labeling efforts in the process. This paper examines the use of transfer learning for efficient multi-view head pose classification with minimal target training data under three challenging situations: (i) where the range of head poses in the source and target images is different, (ii) where source images capture a stationary person while target images capture a moving person whose facial appearance varies under motion due to changing perspective, scale and (iii) a combination of (i) and (ii). On the whole, the presented methods represent novel transfer learning solutions employed in the context of multi-view head pose classification. We demonstrate that the proposed solutions considerably outperform the state-of-the-art through extensive experimental validation. Finally, the DPOSE dataset compiled for benchmarking head pose classification performance with moving persons, and to aid behavioral understanding applications is presented in this work.

Metal-metal charge transfer and interfacial charge transfer mechanism for the visible light photocatalytic activity of cerium and nitrogen co-doped TiO2

Nanosized cerium and nitrogen co-doped TiO2 (Ce-TiO2-xNx) was synthesized by sol gel method and characterized by powder X-ray diffraction (PXRD), X-ray photoelectron spectroscopy (XPS), FESEM, Fourier transform infrared, N-2 adsorption and desorption methods, photoluminescence and ultraviolet-visible (UV-vis) DRS techniques. PXRD analysis shows the dopant decreases the crystallite sizes and slows the crystallization of the titania matrix. XPS confirm the existence of cerium ion in +3 or +4 state, and nitrogen in -3 state in Ce-TiO2-xNx. The modified surface of TiO2 provides highly active sites for the dyes at the periphery of the Ce-O-Ti interface and also inhibits Ce particles from sintering. UV-visible DRS studies show that the metal-metal charge transfer (MMCT) of Ti/Ce assembly (Ti4+/Ce3+ -> Ti3+/Ce4+) is responsible for the visible light photocatalytic activity. Photoluminescence was used to determine the effect of cerium ion on the electron-hole pair separation between the two interfaces Ce-TiO2-xNx and Ce2O3. This separation increases with the increase of cerium and nitrogen ion concentrations of doped samples. The degradation kinetics of methylene blue and methyl violet dyes in the presence of sol gel TiO2, Ce-TiO2-xNx and commercial Degussa P25 was determined. The higher visible light activity of Ce-TiO2-xNx was due to the participation of MMCT and interfacial charge transfer mechanism.

Estimation of the Hall-Petch strengthening coefficient of steels through nanoindentation

A method to estimate the Hall-Petch coefficient k for yield strength and flow stress of steels through nanoindentation experiments is proposed. While determination of k(f) for flow stress is on the basis of grain boundary strengthening evaluated by sharp indentation, k(y) for yield strength was computed with pop-in data from spherical indentations. Good agreement between estimated and literature data, obtained from the tensile tests, validates the proposed methodology. (C) 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Discerning heat transfer in building materials

The function of a building is to ensure safety and thermal comfort for healthy living conditions. Buildings primarily comprise an envelope, which acts as an interface separating the external environment from the indoors environment. The building envelope is primarily responsible for regulating indoor thermal comfort in response to external climatic conditions. It usually comprises a configuration of building materials to thus far provide requisite structural performance. However, studies into building-envelope configurations to provide a particular thermal performance are limited. As the building envelope is exposed to the external environment there will be heat and moisture transfer to the indoor environment through it. The overall phenomenon of heat and moisture transfer depends on the microstructure and configuration within the building material. Further, thermal property of a material is generally dependent on its microstructure, which comprises a network of pores and particles arranged in a definite structure. Thermal behaviour of a building material thus depends on the thermal conductivities of the solid particles, pore micro-structure and its constituent fluid (air and/or moisture). The thermal response of a building envelope is determined by the thermal characteristics of the individual building materials and its configuration. Understanding the heat transfer influenced by the complex networks of pores and particles is a relatively new study in the area of building climatic-response. The current study reviews the heat-transfer mechanisms that determine the thermal performance of a building material attributed to its micro-structure. A theoretical basis for the same is being evolved and its relevance in regulating heat-transfer through building envelopes, walls in particular, is reviewed in this paper. (C) 2014 N.C. Balaji. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Synthesis of mixed glycosyl disulfides/selenenylsulfides using benzyltriethylammonium tetrathiomolybdate as a sulfur transfer reagent

An easy and mild method has been developed for the synthesis of mixed glycosyl disulfides/selenenylsulfides from glycosyl halides and diaryl/dialkyl dichalcogenides in the presence of benzyltriethylammonium tetrathiomolybdate (BnEt3N)(2)MoS4]. The salient feature of this method is the sulfur transfer from BnEt3N](2)MoS4 to form glycosyl disulfides which with excess tetrathiomolybdate further undergo exchange reaction with other dichalcogenides in a one-pot operation. (C) 2014 Elsevier Ltd. All rights reserved.

Investigations on pressure and thickness profiles in carbon fiber-reinforced polymers during vacuum assisted resin transfer molding

Systematic experiments have been carried out by monitoring the in-situ pressure and thickness profiles for three different configurations, viz., flat plate, flat plate with a central circular hole, and an L-section using vacuum assisted resin transfer molding (VARTM) process. The effect of anisotropy on resin flow has been quantified by considering uni-directional carbon fiber preforms with 0 degrees and 90 degrees orientation to the flow direction for each configuration. A quasi-isotropic 45 degrees/0 degrees/-45 degrees/90 degrees](S) layup has also been included for flat plate case. Additionally, the study has been extended to understand the effect of using high permeability medium for each configuration. Fluid pressure profiles and thickness variation profiles have been obtained using an array of pressure sensors and linear variable differential transformers for each configuration. Experimental data reveal that anisotropy (due to changing fiber orientations), configuration, and gravity significantly change fluid pressure and displacement fields obtained during VARTM.

Heterogeneity in Dye-TiO2 Interactions Dictate Charge Transfer Efficiencies for Diketopyrrolopyrrole-Based Polymer Sensitized Solar Cells

Power conversion efficiency of a solar cell is a complex parameter which usually hides the molecular details of the charge generation process. For rationally tailoring the overall device efficiency of the dye-sensitized solar cell, detailed molecular understanding of photoinduced reactions at the dye-TiO2 interface has to be achieved. Recently, near-IR absorbing diketopyrrolopyrrole-based (DPP) low bandgap polymeric dyes with enhanced photostabilities have been used for TiO2 sensitization with moderate efficiencies. To improve the reported device performances, a critical analysis of the polymerTiO(2) interaction and electron transfer dynamics is imperative. Employing a combination of time-resolved optical measurements complemented by low temperature EPR and steady-state Raman spectroscopy on polymerTiO(2) conjugates, we provide direct evidence for photoinduced electron injection from the TDPP-BBT polymer singlet state into TiO2 through the C-O group of the DPP-core. A detailed excited state description of the electron transfer process in films reveals instrument response function (IRF) limited (<110 fs) charge injection from a minor polymer fraction followed by a picosecond recombination. The major fraction of photoexcited polymers, however, does not show injection indicating pronounced ground state heterogeneity induced due to nonspecific polymerTiO(2) interactions. Our work therefore underscores the importance of gathering molecular-level insight into the competitive pathways of ultrafast charge generation along with probing the chemical heterogeneity at the nanoscale within the polymerTiO2 films for optimizing photovoltaic device efficiencies.

Large carbon cluster thin film gauges for measuring aerodynamic heat transfer rates in hypersonic shock tunnels

Different types of Large Carbon Cluster (LCC) layers are synthesized by a single-step pyrolysis technique at various ratios of precursor mixture. The aim is to develop a fast responsive and stable thermal gauge based on a LCC layer which has relatively good electrical conduction in order to use it in the hypersonic flow field. The thermoelectric property of the LCC layer has been studied. It is found that these carbon clusters are sensitive to temperature changes. Therefore suitable thermal gauges were developed for blunt cone bodies and were tested in hypersonic shock tunnels at a flow Mach number of 6.8 to measure aerodynamic heating. The LCC layer of this thermal gauge encounters high shear forces and a hostile environment for test duration in the range of a millisecond. The results are favorable to use large carbon clusters as a better sensor than a conventional platinum thin film gauge in view of fast responsiveness and stability.

Influence of dynamic operating conditions on the performance of metal hydride based solid sorption cooling systems

The performance of metal hydride based solid sorption cooling systems depends on the driving pressure differential, and the rate of hydrogen transfer between coupled metal hydride beds during cooling and regeneration processes. Conventionally, the mid-plateau pressure difference obtained from `static' equilibrium PCT data are used for the thermodynamic analysis. It is well known that the processes are `dynamic' because the pressure and temperature, and hence the concentration of the hydride beds, are continuously changing. Keeping this in mind, the pair of La0.9Ce0.1Ni5 - LaNi4.7Al0.3 metal hydrides suitable for solid sorption cooling systems were characterised using both static and dynamic methods. It was found that the PCT characteristics, and the resulting enthalpy (Delta H) and entropy (Delta S) values, were significantly different for static and dynamic modes of measurements. In the present study, the solid sorption metal hydride cooling system is analysed taking in to account the actual variation in the pressure difference (Delta P) and the dynamic enthalpy values. Compared to `static' property based analysis, significant decrease in the driving potentials and transferrable amounts of hydrogen, leading to decrease in cooling capacity by 57.8% and coefficient of performance by 31.9% are observed when dynamic PCT data at the flow rate of 80 ml/min are considered. Copyright 2014 (C) Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

Analysis of Entropy Generation During Mixed Convective Heat Transfer of Nanofluids Past a Rotating Circular Cylinder

The entropy generation due to mixed convective heat transfer of nanofluids past a rotating circular cylinder placed in a uniform cross stream is investigated via streamline upwind Petrov-Galerkin based finite element method. Nanosized copper (Cu) particles suspended in water are used with Prandtl number (Pr)=6.9. The computations are carried out at a representative Reynolds number (Re) of 100. The dimensionless cylinder rotation rate, a, is varied between 0 and 2. The range of nanoparticle volume fractions (phi) considered is 0 <= phi <= 5%. Effect of aiding buoyancy is brought about by considering two fixed values of the Richardson number (Ri) as 0.5 and 1.0. A new model for predicting the effective viscosity and thermal conductivity of dilute suspensions of nanoscale colloidal particles is presented. The model addresses the details of the agglomeration-deagglomeration in tune with the pertinent variations in the effective particulate dimensions, volume fractions, as well as the aggregate structure of the particulate system. The total entropy generation is found to decrease sharply with cylinder rotation rates and nanoparticle volume fractions. Increase in nanoparticle agglomeration shows decrease in heat transfer irreversibility. The Bejan number falls sharply with increase in alpha and phi.

Charge Transfer Induces Formation of Stimuli-Responsive, Chiral, Cohesive Vesicles-on-a-String that Eventually Turn into a Hydrogel

A charge transfer (CT) mediated two-component, multistimuli responsive supergelation involving a L-histidine-appended pyrenyl derivative (PyHisOMe) as a donor and an asymmetric bolaamphiphilic naphthalene-diimide (Asym-NDI) derivative as an acceptor in a 2: 1 mixture of H2O/MeOH was investigated. Asym-NDI alone self-assembled into pH-responsive vesicular nanostructures in water. Excellent selectivity in CT gel formation was achieved in terms of choosing amino acid appended pyrenyl donor scaffolds. Circular di-chroism and morphological studies suggested formation of chiral, interconnected vesicular assemblies resembling ``pearls-on-a-string'' from these CT mixed stacks. XRD studies revealed the formation of monolayer lipid membranes from these CT mixed stacks that eventually led to the formation of individual vesicles. Strong cohesive forces among the interconnected vesicles originate from the protrusion of the oxyethylene chains from the surfaces of the chiral vesicles.

Structural, thermal and quantum chemical studies of p-coumaric and caffeic acids

Two hydroxycinnamic acids viz., p-coumaric, and caffeic acids have been extracted and purified from Parthenium hysterophorus, subsequently characterized by elemental analysis, FT-IR, NMR, single crystal X-ray crystallography. The optimized structures of these acids were calculated in terms of density functional theory by Gaussian 09. The validation of experimental and theoretically obtained data for structural parameters such as bond lengths and bond angles has have been carried out to analyze the statistical significance by curve fitting analysis and the values of correlation coefficient found to be 0.985, 0.992, and 0.984, 0.975 in p-coumaric, and caffeic acids, respectively. The calculated HOMO and LUMO energies show the eventual charge transfer interaction within the molecule. Thermal studies were also carried out by thermogravimetry (TG), differential thermogravimetric analysis (DTA), and derivative thermogravimetry (DTG). (C) 2014 Elsevier B.V. All rights reserved.

Electron transfer statistics and thermal fluctuations in molecular junctions

We derive analytical expressions for probability distribution function (PDF) for electron transport in a simple model of quantum junction in presence of thermal fluctuations. Our approach is based on the large deviation theory combined with the generating function method. For large number of electrons transferred, the PDF is found to decay exponentially in the tails with different rates due to applied bias. This asymmetry in the PDF is related to the fluctuation theorem. Statistics of fluctuations are analyzed in terms of the Fano factor. Thermal fluctuations play a quantitative role in determining the statistics of electron transfer; they tend to suppress the average current while enhancing the fluctuations in particle transfer. This gives rise to both bunching and antibunching phenomena as determined by the Fano factor. The thermal fluctuations and shot noise compete with each other and determine the net (effective) statistics of particle transfer. Exact analytical expression is obtained for delay time distribution. The optimal values of the delay time between successive electron transfers can be lowered below the corresponding shot noise values by tuning the thermal effects. (C) 2015 AIP Publishing LLC.

Comparative evaluation of inversion approaches of the radiative transfer model for estimation of crop biophysical parameters

The inversion of canopy reflectance models is widely used for the retrieval of vegetation properties from remote sensing. This study evaluates the retrieval of soybean biophysical variables of leaf area index, leaf chlorophyll content, canopy chlorophyll content, and equivalent leaf water thickness from proximal reflectance data integrated broadbands corresponding to moderate resolution imaging spectroradiometer, thematic mapper, and linear imaging self scanning sensors through inversion of the canopy radiative transfer model, PROSAIL. Three different inversion approaches namely the look-up table, genetic algorithm, and artificial neural network were used and performances were evaluated. Application of the genetic algorithm for crop parameter retrieval is a new attempt among the variety of optimization problems in remote sensing which have been successfully demonstrated in the present study. Its performance was as good as that of the look-up table approach and the artificial neural network was a poor performer. The general order of estimation accuracy for para-meters irrespective of inversion approaches was leaf area index > canopy chlorophyll content > leaf chlorophyll content > equivalent leaf water thickness. Performance of inversion was comparable for broadband reflectances of all three sensors in the optical region with insignificant differences in estimation accuracy among them.

Standardizing Patent Data Cleaning in a University Technology Transfer Office

There are multiple goals of a technology transfer office (TTO) based in a university system. Whilst commercialization is a critical goal, maintenance and cleaning of the TTO's database needs detailing. Literature in the area is scarce and only some researchers make reference to TTO data cleaning. During an attempt to understand the commercial strategy of a university TTO in Bangalore the challenge of data cleaning was encountered. This paper describes a case study of data cleaning at an Indian university based TTO. 382 patent records were analyzed in the study. The case study first describes the back ground of the university system. Second, the method to clean the data and the experiences encountered are highlighted. Insights drawn indicate that patent data cleaning in a TTO is a specialized area which needs attention. Overlooking this activity can have legal implications and may result in an inability to commercialize the patent. Two levels of patent data cleaning are discussed in this case study. Best practices of data cleaning in academic TTOs are discussed.