arXiv subjects

Bo Zhao

Publications and source records attributed to Bo Zhao.

At least 289 records · Page 16Linked to original sources

arXiv2019

Axion-Field-Enabled Nonreciprocal Thermal Radiation in Weyl Semimetals

Objects around us constantly emit and absorb thermal radiation. The emission and absorption processes are governed by two fundamental radiative properties: emissivity and absorptivity.For reciprocal systems, the emissivity and absorptivity are restricted to be equal by Kirchhoff's law of thermal radiation. This restriction limits the degree of freedom to control thermal radiation and contributes to an intrinsic loss mechanism in photonic energy harvesting systems such assolar cells. Existing approaches to violate Kirchhoff's law typically utilize conventional magneto-optical effects in the presence of an external magnetic field. However, these approaches require either a strong magnetic field (~3T), or narrow-band resonances under a moderate magnetic field (~0.3T), because the non-reciprocity in conventional magneto-optical effects isusually weak in the thermal wavelength range. Here, we show that the axion electrodynamics in magnetic Weyl semimetals can be used to construct strongly nonreciprocal thermal emitters that near completely violate Kirchhoff's law over broad angular and frequency ranges, without requiring any external magnetic field. The non-reciprocity moreover is strongly temperature tunable, opening new possibilities for active non-reciprocal devices in controlling thermal radiation.

physics.optics↗

arXiv2019

Nonreciprocal radiative heat transfer between two planar bodies

We develop an analytical framework for nonreciprocal radiative heat transfer in two-body planar systems. Based on our formalism, we identify effects that are uniquely nonreciprocal in near-field heat transfer in planar systems. We further introduce a general thermodynamic constraint that is applicable for both reciprocal and nonreciprocal planar systems, in agreement with the second law of thermodynamics. We numerically demonstrate our findings in an example system consisting of magneto-optical materials. Our formalism applies to both near- and far-field regimes, opening opportunities for exploiting nonreciprocity in two-body radiative heat transfer systems.

physics.optics↗

arXiv2019

LabPipe: an extensible informatics platform to streamline management of metabolomics data and metadata

Summary: Data management in clinical metabolomics studies is often inadequate. To improve this situation we created LabPipe to provide a guided, customisable approach to study-specific sample collection. It is driven through a local client which manages the process and pushes local data to a remote server through an access controlled web API. The platform is able to support data management for different sampling approaches across multiple sites / studies and is now an essential study management component for supporting clinical metabolomics locally at the EPSRC/MRC funded East Midlands Breathomics Pathology Node. Availability and Implementation: LabPipe is freely available to download under a non-commercial open-source license (NPOSL 3.0) along with documentation and installation instructions at http://labpipe.org. Contact: rob.free@le.ac.uk

cs.CE↗

arXiv2019

DwNet: Dense warp-based network for pose-guided human video generation

Generation of realistic high-resolution videos of human subjects is a challenging and important task in computer vision. In this paper, we focus on human motion transfer - generation of a video depicting a particular subject, observed in a single image, performing a series of motions exemplified by an auxiliary (driving) video. Our GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the required subject appearance, in the form of the texture, from a source image into a desired pose. Temporal consistency is maintained by further conditioning the decoding process within a GAN on the previously generated frame. In this way a video is generated in an iterative and recurrent fashion. We illustrate the efficacy of our approach by showing state-of-the-art quantitative and qualitative performance on two benchmark datasets: TaiChi and Fashion Modeling. The latter is collected by us and will be made publicly available to the community.

cs.CV↗

arXiv2019

Image Generation from Layout

Despite significant recent progress on generative models, controlled generation of images depicting multiple and complex object layouts is still a difficult problem. Among the core challenges are the diversity of appearance a given object may possess and, as a result, exponential set of images consistent with a specified layout. To address these challenges, we propose a novel approach for layout-based image generation; we call it Layout2Im. Given the coarse spatial layout (bounding boxes + object categories), our model can generate a set of realistic images which have the correct objects in the desired locations. The representation of each object is disentangled into a specified/certain part (category) and an unspecified/uncertain part (appearance). The category is encoded using a word embedding and the appearance is distilled into a low-dimensional vector sampled from a normal distribution. Individual object representations are composed together using convolutional LSTM, to obtain an encoding of the complete layout, and then decoded to an image. Several loss terms are introduced to encourage accurate and diverse generation. The proposed Layout2Im model significantly outperforms the previous state of the art, boosting the best reported inception score by 24.66% and 28.57% on the very challenging COCO-Stuff and Visual Genome datasets, respectively. Extensive experiments also demonstrate our method's ability to generate complex and diverse images with multiple objects.

cs.CV↗

arXiv2019

Gas flow and accretion via spiral streamers and circumstellar disks in a young binary protostar

The majority of stars are part of gravitationally bound stellar systems, such as binaries. Observations of protobinary systems constrain the conditions that lead to stellar multiplicity and subsequent orbital evolution. We report high-angular resolution observations of the circumbinary disk around [BHB2007] 11, a young binary protostar system. The two protostars are embedded in circumstellar disks that have radii of 2 to 3 astronomical units and probably contain a few Jupiter masses. These systems are surrounded by a complex structure of filaments connecting to the larger circumbinary disk. We also observe accretion and radio jets associated with the protobinary system. The accretion is preferentially onto the lower-mass protostar, consistent with theoretical predictions.

astro-ph.SR↗

arXiv2019

Broadening near-field emission for performance enhancement in thermophotovoltaics

The conventional notion for achieving high efficiency in thermophotovoltaics (TPVs) is to use a monochromatic emission at a photon energy corresponding to the band gap of the cell. Here, we prove theoretically that such a notion is only accurate under idealized conditions, and further show that when non-radiative recombination is taken into account, efficiency improvement can be achieved by broadening the emission spectrum, due to an enhancement in the open-circuit voltage. Broadening the emission spectrum also improves the electrical power density, by increasing the short-circuit current. To practically illustrate these findings, we focus on surface polariton-mediated near-field TPVs. We propose a versatile design strategy for broadening the emission spectrum via stacking of multiple plasmonic thin film layers. As an example, we consider a realistic ITO/InAs TPV, and predict a conversion efficiency of $50\%$ simultaneously with a power density of nearly $80$ W$/\mathrm{cm}^2$ at a $1300$ K emitter temperature. The performance of our proposed system far exceeds previous works in similar systems using a single plasmonic layer emitter.

physics.optics↗

arXiv2019

Disk Formation in Magnetized Dense Cores with Turbulence and Ambipolar Diffusion

Disks are essential to the formation of both stars and planets, but how they form in magnetized molecular cloud cores remains debated. This work focuses on how the disk formation is affected by turbulence and ambipolar diffusion (AD), both separately and in combination, with an emphasis on the protostellar mass accretion phase of star formation. We find that a relatively strong, sonic turbulence on the core scale strongly warps but does not completely disrupt the well-known magnetically-induced flattened pseudodisk that dominates the inner protostellar accretion flow in the laminar case, in agreement with previous work. The turbulence enables the formation of a relatively large disk at early times with or without ambipolar diffusion, but such a disk remains strongly magnetized and does not persist to the end of our simulation unless a relatively strong ambipolar diffusion is also present. The AD-enabled disks in laminar simulations tend to fragment gravitationally. The disk fragmentation is suppressed by initial turbulence. The ambipolar diffusion facilitates the disk formation and survival by reducing the field strength in the circumstellar region through magnetic flux redistribution and by making the field lines there less pinched azimuthally, especially at late times. We conclude that turbulence and ambipolar diffusion complement each other in promoting disk formation. The disks formed in our simulations inherit a rather strong magnetic field from its parental core, with a typical plasma-$β$ of order a few tens or smaller, which is 2-3 orders of magnitude lower than the values commonly adopted in MHD simulations of protoplanetary disks. To resolve this potential tension, longer-term simulations of disk formation and evolution with increasingly more realistic physics are needed.

astro-ph.SR↗

arXiv2019

The specific angular momentum radial profile in dense cores: improved initial conditions for disk formation

The determination of the specific angular momentum radial profile, $j(r)$, in the early stages of star formation is crucial to constrain star and circumstellar disk formation theories. The specific angular momentum is directly related to the largest Keplerian disk possible, and it could constrain the angular momentum removal mechanism. We determine $j(r)$ towards two Class 0 objects and a first hydrostatic core candidate in the Perseus cloud, which is consistent across all three sources and well fit with a single power-law relation between 800 and 10,000\,au: $j_{fit}(r)=10^{-3.60\pm0.15}\left(r/\textrm{1,000au}\right)^{1.80\pm0.04}$ km s$^{-1}$ pc. This power-law relation is in between solid body rotation ($\propto r^2$) and pure turbulence ($\propto r^{1.5}$). This strongly suggests that even at 1,000\,au, the influence of the dense core's initial level of turbulence or the connection between core and the molecular cloud is still present. The specific angular momentum at 10,000\,au is $\approx3\times$ higher than previously estimated, while at 1,000\,au it is lower by $2\times$. We do not find a region of conserved specific angular momentum, although it could still be present at a smaller radius. We estimate an upper limit to the largest Keplerian disk radius of 60\,au, which is small but consistent with published upper limits. Finally, these results suggest that more realistic initial conditions for numerical simulations of disk formation are needed. Some possible solutions include: a) use a larger simulation box to include some level of driven turbulence or connection to the parental cloud, or b) incorporate the observed $j(r)$ to setup the dense core kinematics initial conditions.

astro-ph.GA↗

arXiv2019

Observation of interference between resonant and detuned STIRAP in the adiabatic creation of $^{23}$Na$^{40}$K molecules

Stimulated Raman adiabatic passage (STIRAP) allows to efficiently transferring the populations between two discrete quantum states and has been used to prepare molecules in their rovibrational ground state. In realistic molecules, a well-resolved intermediate state is usually selected to implement the resonant STIRAP. Due to the complex molecular level structures, the detuned STIRAP always coexists with the resonant STIRAP and may cause unexpected interference phenomenon. However, it is generally accepted that the detuned STIRAP can be neglected if compared with the resonant STIRAP. Here we report on the first observation of interference between the resonant and detuned STIRAP in the adiabatic creation of $^{23}$Na$^{40}$K ground-state molecules. The interference is identified by observing that the number of Feshbach molecules after a round-trip STIRAP oscillates as a function of the hold time, with a visibility of about 90\%. This occurs even if the intermediate excited states are well resolved, and the single-photon detuning of the detuned STIRAP is about one order of magnitude larger than the linewidth of the excited state and the Rabi frequencies of the STIRAP lasers. Moreover, the observed interference indicates that if more than one hyperfine level of the ground state is populated, the STIRAP prepares a coherent superposition state among them, but not an incoherent mixed state. Further, the purity of the hyperfine levels of the created ground state can be quantitatively determined by the visibility of the oscillation.

physics.atom-ph↗

arXiv2019

Interpretable Spatio-temporal Attention for Video Action Recognition

Inspired by the observation that humans are able to process videos efficiently by only paying attention where and when it is needed, we propose an interpretable and easy plug-in spatial-temporal attention mechanism for video action recognition. For spatial attention, we learn a saliency mask to allow the model to focus on the most salient parts of the feature maps. For temporal attention, we employ a convolutional LSTM based attention mechanism to identify the most relevant frames from an input video. Further, we propose a set of regularizers to ensure that our attention mechanism attends to coherent regions in space and time. Our model not only improves video action recognition accuracy, but also localizes discriminative regions both spatially and temporally, despite being trained in a weakly-supervised manner with only classification labels (no bounding box labels or time frame temporal labels). We evaluate our approach on several public video action recognition datasets with ablation studies. Furthermore, we quantitatively and qualitatively evaluate our model's ability to localize discriminative regions spatially and critical frames temporally. Experimental results demonstrate the efficacy of our approach, showing superior or comparable accuracy with the state-of-the-art methods while increasing model interpretability.

cs.CV↗

arXiv2019

A Survey on Matrix Completion: Perspective of Signal Processing

Matrix completion (MC) is a promising technique which is able to recover an intact matrix with low-rank property from sub-sampled/incomplete data. Its application varies from computer vision, signal processing to wireless network, and thereby receives much attention in the past several years. There are plenty of works addressing the behaviors and applications of MC methodologies. This work provides a comprehensive review for MC approaches from the perspective of signal processing. In particular, the MC problem is first grouped into six optimization problems to help readers understand MC algorithms. Next, four representative types of optimization algorithms solving the MC problem are reviewed. Ultimately, three different application fields of MC are described and evaluated.

eess.SP↗

arXiv2019

Multiple Aging Mechanisms in Ferroelectric Deuterated Potassium Dihydrogen Phosphate

The anomalously large dielectric aging in ferroelectric partially deuterated potassium dihydrogen phosphate (DKDP) is found to have multiple distinct mechanisms. Two components cause decreases in dielectric response over a limited range of fields around the aging field. A large fraction of this aging occurs on time scales of ~1000s after a field change, as expected for a hydrogen/deuterium diffusion mechanism. A slower component can give almost complete loss of domain-wall dielectric response at the aging field after weeks of aging. There is also a particularly unusual aging in which the dielectric response increases with time after rapid cooling.

cond-mat.mtrl-sci↗

arXiv2019

The central 1000 AU of a pre-stellar core revealed with ALMA. I. 1.3 mm continuum observations

Stars like our Sun form in self-gravitating dense and cold structures within interstellar clouds, called pre-stellar cores. Although much is known about the physical structure of dense clouds just before and soon after the switch-on of a protostar, the central few thousand astronomical units (au) of pre-stellar cores are unexplored. It is within these central regions that stellar systems assemble and fragmentation may take place, with the consequent formation of binaries and multiple systems. We present ALMA Band 6 observations (ACA and 12m array) of the dust continuum emission of the 8 Msun pre-stellar core L1544, with angular resolution of 2'' x 1.6'' (linear resolution 270 au x 216 au). Within the primary beam, a compact region of 0.1 Msun, which we call a "kernel", has been unveiled. The kernel is elongated, with a central flat zone with radius Rker ~ 10'' (~ 1400 au). The average number density within Rker is ~1 x 10^6 cm^{-3}, with possible local density enhancements. The region within Rker appears to have fragmented, but detailed analysis shows that similar substructure can be reproduced by synthetic interferometric observations of a smooth centrally concentrated dense core with a similar central flat zone. The presence of a smooth kernel within a dense core is in agreement with non-ideal magneto-hydro-dynamical simulations of a contracting cloud core with a peak number density of 1 x 10^7 cm^{-3}. Dense cores with lower central densities are completely filtered out when simulated 12m-array observations are carried out. These observations demonstrate that the kernel of dynamically evolved dense cores can be investigated at high angular resolution with ALMA.

astro-ph.SR↗

arXiv2019

JCMT POL-2 and ALMA polarimetric observations of 6000-100 au scales in the protostar B335: linking magnetic field and gas kinematics in observations and MHD simulations

We present our analysis of the magnetic field structures from 6000 au to 100 au scales in the Class 0 protostar B335 inferred from our JCMT POL-2 observations and the ALMA archival polarimetric data. To interpret the observational results, we perform a series of (non-)ideal MHD simulations of the collapse of a rotating non-turbulent dense core, whose initial conditions are adopted to be the same as observed in B335, and generate synthetic polarization maps. The comparison of our JCMT and simulation results suggests that the magnetic field on a 6000 au scale in B335 is pinched and well aligned with the bipolar outflow along the east-west direction. Among all our simulations, the ALMA polarimetric results are best explained with weak magnetic field models having an initial mass-to-flux ratio of 9.6. However, we find that with the weak magnetic field, the rotational velocity on a 100 au scale and the disk size in our simulations are larger than the observational estimates by a factor of several. An independent comparison of our simulations and the gas kinematics in B335 observed with the SMA and ALMA favors strong magnetic field models with an initial mass-to-flux ratio smaller than 4.8. We discuss two possibilities resulting in the different magnetic field strengths inferred from the polarimetric and molecular-line observations, (1) overestimated rotational-to-gravitational energy in B335 and (2) additional contributions in the polarized intensity due to scattering on a 100 au scale.

astro-ph.SR↗

arXiv2018

High reflection from a one-dimensional array of graphene nanoribbons

We show that up to 90% reflectivity can be achieved by using guided plasmonic resonances in a one-dimensional periodic array of plasmonic nanoribbon. In general, to achieve strong reflection from a guided resonance system requires one to operate in the strongly over-coupled regime where the radiative decay rate dominates over the intrinsic loss rate of the resonances. Using an argument similar to what has been previous used to derive the Chu-Harrington limit for antennas, we show theoretically that there is no intrinsic limit for the radiative decay rate even when the system has an atomic scale thickness, in contrast to the existence of such limits on antennas. We also show that the current distribution due to plasmonic resonance can be designed to achieve very high external radiative rate. Our results show that high reflectivity can be achieved in an atomically-thin graphene layer, pointing to a new opportunity for creating atomically-thin optical devices.

physics.optics↗

arXiv2018

Gate-tunable near-field heat transfer

Active control over the flow of heat in the near-field holds promise for nanoscale thermal management, with applications in refrigeration, thermophotovoltaics, and thermal circuitry. Analogously to its electronic counterpart, the metal-oxide-semiconductor (MOS) capacitor, we propose a thermal switching mechanism based on accumulation and depletion of charge carriers in an ultra-thin plasmonic film, via application of external bias. In our proposed configuration, the plasmonic film is placed on top of a polaritonic dielectric material that provides a surface phonon polariton (SPhP) thermal channel, while also ensuring electrical insulation for application of large electric fields. The variation of carrier density in the plasmonic film enables the control of the surface plasmon polariton (SPP) thermal channel. We show that the interaction of the SPP with the SPhP significantly enhances the net heat transfer. We study SiC as the oxide and explore three classes of gate-tunable plasmonic materials: transparent conductive oxides, doped semiconductors, and graphene, and theoretically predict contrast ratios as high as 225%.

physics.app-ph↗

arXiv2018

Optimal Experiment Design for Magnetic Resonance Fingerprinting: Cramér-Rao Bound Meets Spin Dynamics

Magnetic resonance (MR) fingerprinting is a new quantitative imaging paradigm, which simultaneously acquires multiple MR tissue parameter maps in a single experiment. In this paper, we present an estimation-theoretic framework to perform experiment design for MR fingerprinting. Specifically, we describe a discrete-time dynamic system to model spin dynamics, and derive an estimation-theoretic bound, i.e., the Cramer-Rao bound (CRB), to characterize the signal-to-noise ratio (SNR) efficiency of an MR fingerprinting experiment. We then formulate an optimal experiment design problem, which determines a sequence of acquisition parameters to encode MR tissue parameters with the maximal SNR efficiency, while respecting the physical constraints and other constraints from the image decoding/reconstruction process. We evaluate the performance of the proposed approach with numerical simulations, phantom experiments, and in vivo experiments. We demonstrate that the optimized experiments substantially reduce data acquisition time and/or improve parameter estimation. For example, the optimized experiments achieve about a factor of two improvement in the accuracy of $T_2$ maps, while keeping similar or slightly better accuracy of $T_1$ maps. Finally, as a remarkable observation, we find that the sequence of optimized acquisition parameters appears to be highly structured rather than randomly/pseudo-randomly varying as is prescribed in the conventional MR fingerprinting experiments.

eess.SP↗