arXiv subjects

Huan Yang

Publications and source records attributed to Huan Yang.

At least 181 records · Page 10Linked to original sources

arXiv2021

Improving Visual Quality of Image Synthesis by A Token-based Generator with Transformers

We present a new perspective of achieving image synthesis by viewing this task as a visual token generation problem. Different from existing paradigms that directly synthesize a full image from a single input (e.g., a latent code), the new formulation enables a flexible local manipulation for different image regions, which makes it possible to learn content-aware and fine-grained style control for image synthesis. Specifically, it takes as input a sequence of latent tokens to predict the visual tokens for synthesizing an image. Under this perspective, we propose a token-based generator (i.e.,TokenGAN). Particularly, the TokenGAN inputs two semantically different visual tokens, i.e., the learned constant content tokens and the style tokens from the latent space. Given a sequence of style tokens, the TokenGAN is able to control the image synthesis by assigning the styles to the content tokens by attention mechanism with a Transformer. We conduct extensive experiments and show that the proposed TokenGAN has achieved state-of-the-art results on several widely-used image synthesis benchmarks, including FFHQ and LSUN CHURCH with different resolutions. In particular, the generator is able to synthesize high-fidelity images with 1024x1024 size, dispensing with convolutions entirely.

cs.CV↗

arXiv2021

Collision-sticking kinetics of acid-base clusters and its influence on atmospheric new particle formation

Kinetics of collision-sticking processes between vapor molecules and molecular clusters of low volatile compounds facilitates the initial steps of atmospheric clustering. Conventional theoretical models are quite inaccurate due to the neglection of long-range interactions that essentially govern the kinetics of these microscopic phenomena. Here, we present a consistent and generic theoretical model for evaluating collision rates between molecules and molecular clusters with intermolecular potentials properly incorporated. The model requires solely the elementary molecule-molecule potential as a-priori information but predicts collision rates of molecular clusters at arbitrary sizes, with an accuracy comparable to all-atom molecular dynamics simulations we performed for sulfuric acid-dimethylamine clusters, a typical example of acid-base induced clustering. The carefully devised simulations validate the theoretical model and elucidate the kinetics of the molecular collision-sticking process. It is found that the vibrational coupling after the collision between a sulfuric acid molecule and a sulfuric acid cluster can be occasionally unsuccessful, i.e., no stable bond is formed after the collision. However, introducing dimethylamine molecules to the sulfuric acid cluster can notably increase the probability of forming stable bonds and hence stable product clusters. The results offer fundamental insights into the initial steps of molecular clustering and will facilitate the development of efficient kinetic approach-based nucleation models.

physics.chem-ph↗

arXiv2021

Orbit Tomography of Binary Supermassive Black Holes with Very Long Baseline Interferometry

In this work, we study how to infer the orbit of a supermassive black hole binary (SMBHB) by time-dependent measurements with Very Long Baseline Interferometry (VLBI), such as the Event Horizon Telescope (EHT). Assuming a point-like luminosity image model, we show that with multiple years of observations by EHT, it is possible to recover the SMBHB orbital parameters -- eccentricity, (rescaled) semi-major axis, orbital frequency, and orbital angles -- from their time-varying visibilities even if the binaries orbital period is a few times longer than the duration of observation. Together with the future gravitational wave detections of resolved sources of SMBHBs with Pulsar Timing Array, and/or the detections of optical-band light curves, we will be able to further measure the individual mass of the binary, and also determine the Hubble constant if the total mass of the binary is measured through the light curves of the two black holes or by alternative methods.

gr-qc↗

arXiv2021

Magnetic Feshbach resonances in collisions of $^{23}$Na$^{40}$K with $^{40}$K

We present measurements of more than 80 magnetic Feshbach resonances in collisions of ultracold $^{23}$Na$^{40}$K with $^{40}$K. We assign quantum numbers to a group of low-field resonances and show that they are due to long-range states of the triatomic complex in which the quantum numbers of the separated atom and molecule are approximately preserved. The resonant states are not members of chaotic bath of short-range states. Similar resonances are expected to be a common feature of alkali-metal diatom + atom systems.

physics.atom-ph↗

arXiv2021

No observation of chiral flux current in the topological kagome metal CsV$_{3}$Sb$_{5}$

Compounds with kagome lattice usually host many exotic quantum states, including the quantum spin liquid, non-trivial topological Dirac bands and a strongly renormalized flat band, etc. Recently an interesting vanadium based kagome family $A$V$_{3}$Sb$_{5}$ ($A$ = K, Rb, or Cs) was discovered, and these materials exhibit multiple interesting properties, including unconventional saddle-point driving charge density wave (CDW) state, superconductivity, etc. Furthermore, some experiments show anomalous Hall effect which inspires that there might be some chiral flux current states. Here we report scanning tunneling measurements by using spin-polarized tips. Although we have observed clearly the $2a_0\times2a_0$ CDW and $4a_0$ stripe orders, the well-designed experiments with refined spin-polarized tips do not reveal any trace of the chiral flux current phase in CsV$_3$Sb$_5$ within the limits of experimental accuracy. No observation of the local magnetic moment in our experiments may put an upper bound constraint on the magnitude of magnetic moments induced by the possible chiral loop current which has a time-reversal symmetry breaking along $c$-axis in CsV$_{3}$Sb$_{5}$.

cond-mat.str-el↗

arXiv2021

The Next Generation Global Gravitational Wave Observatory: The Science Book

The next generation of ground-based gravitational-wave detectors will observe coalescences of black holes and neutron stars throughout the cosmos, thousands of them with exceptional fidelity. The Science Book is the result of a 3-year effort to study the science capabilities of networks of next generation detectors. Such networks would make it possible to address unsolved problems in numerous areas of physics and astronomy, from Cosmology to Beyond the Standard Model of particle physics, and how they could provide insights into workings of strongly gravitating systems, astrophysics of compact objects and the nature of dense matter. It is inevitable that observatories of such depth and finesse will make new discoveries inaccessible to other windows of observation. In addition to laying out the rich science potential of the next generation of detectors, this report provides specific science targets in five different areas in physics and astronomy and the sensitivity requirements to accomplish those science goals. This report is the second in a six part series of reports by the GWIC 3G Subcommittee: i) Expanding the Reach of Gravitational Wave Observatories to the Edge of the Universe, ii) The Next Generation Global Gravitational Wave Observatory: The Science Book (this report), iii) 3G R&D: R&D for the Next Generation of Ground-based Gravitational Wave Detectors, iv) Gravitational Wave Data Analysis: Computing Challenges in the 3G Era, v) Future Ground-based Gravitational-wave Observatories: Synergies with Other Scientific Communities, and vi) An Exploration of Possible Governance Models for the Future Global Gravitational-Wave Observatory Network.

gr-qc↗

arXiv2021

New spectroscopic confirmations of Lyman-$α$ emitters at z $\sim$ 7 from the LAGER survey

We report spectroscopic confirmations of 15 Lyman-alpha galaxies at $z\sim7$, implying a spectroscopic confirmation rate of $\sim$80% on candidates selected from LAGER (Lyman-Alpha Galaxies in the Epoch of Reionization), which is the largest (24 deg$^2$) survey aimed at finding Lyman-alpha emitters (LAEs) at $z\sim7$ using deep narrow-band imaging from DECam at CTIO. LAEs at high-redshifts are sensitive probes of cosmic reionization and narrow-band imaging is a robust and effective method for selecting a large number of LAEs. In this work, we present results from the spectroscopic follow-up of LAE candidates in two LAGER fields, COSMOS and WIDE-12, using observations from Keck/LRIS. We report the successful detection of Ly$α$ emission in 15 candidates (11 in COSMOS and 4 in WIDE-12 fields). Three of these in COSMOS have matching confirmations from a previous LAGER spectroscopic follow-up and are part of the overdense region, LAGER-$z7$OD1. Additionally, two candidates that were not detected in the LRIS observations have prior spectroscopic confirmations from Magellan. Including these, we obtain a spectroscopic confirmation success rate of $\sim$$80$% for LAGER LAE candidates. Apart from Ly$α$, we do not detect any other UV nebular lines in our LRIS spectra; however, we estimate a 2$σ$ upper limit for the ratio of NV/Ly$α$, $f_{NV}/f_{Lyα} \lesssim 0.27$, which implies that ionizing emission from these sources is mostly dominated by star formation. Including confirmations from this work, a total of 33 LAE sources from LAGER are now spectroscopically confirmed. LAGER has more than doubled the sample of spectroscopically confirmed LAE sources at $z\sim7$.

astro-ph.GA↗

arXiv2021

Mimicking Kerr's multipole moments

Multipole moments carry a lot of information about the gravitational field. Nonetheless, knowing all the multipole moments of an object does not determine conclusively the nature of the object itself. In particular, the multipole moments of the Kerr spacetime are not unique. Here we construct several physically motivated Newtonian objects with multipole moments identical to those of Kerr. Moreover, we also provide a description of how to include post-Newtonian corrections to these objects without changing their multipole moments.

gr-qc↗

arXiv2021

First Constraints on Nuclear Coupling of Axionlike Particles from the Binary Neutron Star Gravitational Wave Event GW170817

Light axion fields, if they exist, can be sourced by neutron stars due to their coupling to nuclear matter, and play a role in binary neutron star mergers. We report on a search for such axions by analysing the gravitational waves from the binary neutron star inspiral GW170817. We find no evidence of axions in the sampled parameter space. The null result allows us to impose constraints on axions with masses below $10^{-11} {\rm eV}$ by excluding the ones with decay constants ranging from $1.6\times10^{16} {\rm GeV}$ to $10^{18} {\rm GeV}$ at $3σ$ confidence level. Our analysis provides the first constraints on axions from neutron star inspirals, and rules out a large region in parameter space that has not been probed by the existing experiments.

hep-ph↗

arXiv2021

Learning Fine-Grained Motion Embedding for Landscape Animation

In this paper we focus on landscape animation, which aims to generate time-lapse videos from a single landscape image. Motion is crucial for landscape animation as it determines how objects move in videos. Existing methods are able to generate appealing videos by learning motion from real time-lapse videos. However, current methods suffer from inaccurate motion generation, which leads to unrealistic video results. To tackle this problem, we propose a model named FGLA to generate high-quality and realistic videos by learning Fine-Grained motion embedding for Landscape Animation. Our model consists of two parts: (1) a motion encoder which embeds time-lapse motion in a fine-grained way. (2) a motion generator which generates realistic motion to animate input images. To train and evaluate on diverse time-lapse videos, we build the largest high-resolution Time-lapse video dataset with Diverse scenes, namely Time-lapse-D, which includes 16,874 video clips with over 10 million frames. Quantitative and qualitative experimental results demonstrate the superiority of our method. In particular, our method achieves relative improvements by 19% on LIPIS and 5.6% on FVD compared with state-of-the-art methods on our dataset. A user study carried out with 700 human subjects shows that our approach visually outperforms existing methods by a large margin.

cs.CV↗

arXiv2021

Wet Extreme Mass Ratio Inspirals May Be More Common For Spaceborne Gravitational Wave Detection

Extreme Mass Ratio Inspirals (EMRIs) can be classified as dry EMRIs and wet EMRIs based on their formation mechanisms. Dry (or the "loss-cone") EMRIs, previously considered as the main EMRI sources for the Laser Interferometer Space Antenna, are primarily produced by multi-body scattering in the nuclear star cluster and gravitational capture. In this work, we highlight an alternative EMRI formation channel: (wet) EMRI formation assisted by the accretion flow around accreting galactic-center massive black holes (MBHs). In this channel, the accretion disk captures stellar-mass black holes that are intially moving on inclined orbits, and subsequently drives them to migrate towards the MBH - this process boosts the formation rate of EMRIs in such galaxies by orders of magnitude. Taking into account the fraction ($\mathcal O(10^{-2}-10^{-1})$) of active galactic nuclei where the MBHs are expected to be rapidly accreting, we forecast that wet EMRIs will contribute an important or even dominant fraction of all detectable EMRIs by spaceborne gravitational wave detectors.

astro-ph.HE↗

arXiv2021

Learning Conditional Knowledge Distillation for Degraded-Reference Image Quality Assessment

An important scenario for image quality assessment (IQA) is to evaluate image restoration (IR) algorithms. The state-of-the-art approaches adopt a full-reference paradigm that compares restored images with their corresponding pristine-quality images. However, pristine-quality images are usually unavailable in blind image restoration tasks and real-world scenarios. In this paper, we propose a practical solution named degraded-reference IQA (DR-IQA), which exploits the inputs of IR models, degraded images, as references. Specifically, we extract reference information from degraded images by distilling knowledge from pristine-quality images. The distillation is achieved through learning a reference space, where various degraded images are encouraged to share the same feature statistics with pristine-quality images. And the reference space is optimized to capture deep image priors that are useful for quality assessment. Note that pristine-quality images are only used during training. Our work provides a powerful and differentiable metric for blind IRs, especially for GAN-based methods. Extensive experiments show that our results can even be close to the performance of full-reference settings.

eess.IV↗

arXiv2021

Domain-Aware Universal Style Transfer

Style transfer aims to reproduce content images with the styles from reference images. Existing universal style transfer methods successfully deliver arbitrary styles to original images either in an artistic or a photo-realistic way. However, the range of 'arbitrary style' defined by existing works is bounded in the particular domain due to their structural limitation. Specifically, the degrees of content preservation and stylization are established according to a predefined target domain. As a result, both photo-realistic and artistic models have difficulty in performing the desired style transfer for the other domain. To overcome this limitation, we propose a unified architecture, Domain-aware Style Transfer Networks (DSTN) that transfer not only the style but also the property of domain (i.e., domainness) from a given reference image. To this end, we design a novel domainness indicator that captures the domainness value from the texture and structural features of reference images. Moreover, we introduce a unified framework with domain-aware skip connection to adaptively transfer the stroke and palette to the input contents guided by the domainness indicator. Our extensive experiments validate that our model produces better qualitative results and outperforms previous methods in terms of proxy metrics on both artistic and photo-realistic stylizations.

cs.CV↗

arXiv2021

Supercritical accretion of stellar-mass compact objects in active galactic nuclei

Accretion disks of active galactic nuclei (AGN) have been proposed as promising sites for producing both (stellar-mass) compact object mergers and extreme mass ratio inspirals. Along with the disk-assisted migration/evolution process, ambient gas materials inevitably accrete onto the compact objects. The description of this process is subject to significant theoretical uncertainties in previous studies. It was commonly assumed that either an Eddington accretion rate or a Bondi accretion rate (or any rate in between) takes place, although these two rates can differ from each other by several orders of magnitude. As a result, the mass and spin evolution of compact objects within AGN disks are essentially unknown. In this work, we construct a relativistic supercritical inflow-outflow model for black hole (BH) accretion. We show that the radiation efficiency of the supercritical accretion of a stellar-mass BH (sBH) is generally too low to explain the proposed electromagnetic counterpart of GW190521. Applying this model to sBHs embedded in AGN disks, we find that, although the gas inflow rates at Bondi radii of these sBHs are in general highly super-Eddington, a large fraction of inflowing gas eventually escapes as outflows so that only a small fraction accretes onto the sBH, resulting in mildly super-Eddington BH absorption in most cases. We also implement this inflow-outflow model to study the evolution of neutron stars (NS) and white dwarfs (WD) in AGN disks, taking into account corrections from star sizes and star magnetic fields. It turns out to be difficult for WDs to grow to the Chandrasekhar limit via accretion because WDs are spun up more efficiently to reach the shedding limit before the Chandrasekhar limit. For NSs the accretion-induced collapse is possible if NS magnetic fields are sufficiently strong, keeping the NS in a slow rotation state during accretion.

astro-ph.HE↗

arXiv2021

Direct Visualization of a Static Incommensurate Antiferromagnetic Order by Suppressing the Superconducting Phase Coherence in Fe-doped Bi2Sr2CaCu2O8+delta

In cuprate superconductors, due to strong electronic correlations, there are multiple intertwined orders which either coexist or compete with superconductivity. Among them the antiferromagnetic (AF) order is the most prominent one. In the region where superconductivity sets in, the long-range AF order is destroyed. Yet the residual short-range AF fluctuations are present up to a much higher doping and their role in the emergence of the superconducting phase is still highly debated. Here, by using a spin polarized scanning tunneling microscope, for the first time, we directly visualize an emergent incommensurate AF order in the nearby region of Fe impurities embedded in the optimally doped Bi2Sr2CaCu2O8+δ (Bi2212). Remarkably the Fe impurities suppress the superconducting coherence peaks with the gapped feature intact, but pin down the ubiquitous short-range incommensurate AF order. Our work shows an intimate relation between antiferromagnetism and superconductivity.

cond-mat.supr-con↗

arXiv2021

Formation Rate of Extreme Mass Ratio Inspirals in Active Galactic Nuclei

Extreme Mass Ratio Inspirals (EMRIs) are important sources for space-borne gravitational wave detectors, such as LISA (Laser Interferometer Space Antenna) and TianQin. Previous EMRI rate studies have focused on the "loss cone" scenario, where stellar-mass black holes (sBHs) are scattered into highly eccentric orbits near the central massive black hole (MBH) via multi-body interaction. In this work, we calculate the rate of EMRIs of an alternative formation channel: EMRI formation assisted by the accretion flow around accreting massive black holes. In this scenario, sBHs and stars on inclined orbits are captured by the accretion disk, and then subsequently migrate towards the MBH, under the influence of density wave generation and head wind. By solving the Fokker-Planck equation incorporating both sBH-sBH/sBH-star scatterings and sBH/star-disk interactions, we find that an accretion disk usually boosts the EMRI formation rate per individual MBH by $\mathcal O(10^1-10^3)$ compared with the canonical "loss cone" formation channel. Taking into account that the fraction of active galactic nucleus (AGNs) is $\sim \mathcal O(10^{-2}-10^{-1})$, where the MBHs are expected to be rapidly accreting, we expect EMRI formation assisted by AGN disks to be an important channel for all EMRIs observed by space-borne gravitational wave detectors. These two channels also predict distinct distributions of EMRI eccentricities and orbit inclinations with respect to the MBH spin equatorial plan, which can be tested by future gravitational wave observations.

astro-ph.HE↗

arXiv2021

Black-Hole Perturbation Plus Post-Newtonian Theory: Hybrid Waveform for Neutron Star Binaries

We consider the motion of nonspinning, compact objects orbiting around a Kerr black hole with tidal couplings. The tide-induced quadrupole moment modifies both the orbital energy and outgoing fluxes, so that over the inspiral timescale there is an accumulative shift in the orbital and gravitational wave phase. Previous studies on compact object tidal effects have been carried out in the Post-Newtonian (PN) and Effective-One-Body (EOB) formalisms. In this work, within the black hole perturbation framework, we propose to characterize the tidal influence in the expansion of mass ratios, while higher-order PN corrections are naturally included. For the equatorial and circular orbit, we derive the leading order, frequency dependent tidal phase shift which agrees with the Post-Newtonian result at low frequencies but deviates at high frequencies. We also find that such phase shift has weak dependence ($\le 10\%$) on the spin of the primary black hole. Combining this black hole perturbation waveform with the Post-Newtonian waveform, we propose a frequency-domain, hybrid waveform that shows comparable accuracy as the EOB waveform in characterizing the tidal effects, as calibrated by numerical relativity simulations. Further improvement is expected as the next-leading order in mass ratio and the higher-PN tidal corrections are included. This hybrid approach is also applicable for generating binary black hole waveforms.

gr-qc↗

arXiv2021

The Compact UV Size of Green Pea Galaxies As Local Analogs of High-redshift Ly$α$-Emitters

We study the dependence of Ly$α$ escape from galaxies on UV continuum size and luminosity using a sample of 40 Green Pea (GP) galaxies, which are the best local analogs of high-redshift Ly$α$-emitters (LAEs). We use the Cosmic Origins Spectrograph near-ultraviolet images from the \textit{Hubble Space Telescope} to measure the UV size and luminosity with $0.047''$ spatial resolution. Like most galaxies the GPs show a log-normal size distribution. They also show a positive correlation between size and UV-continuum luminosity. The slope of the size-continuum luminosity relation for GPs is consistent with those of continuum-selected star-forming galaxies at low and high redshifts. A distinctive feature of GPs is a very compact typical radius of 0.33 kpc with a population spread (1$σ$) of 0.19 kpc. The peak of the size distribution and the intercept of the size-luminosity relation of GPs are noticeably smaller than those of continuum-selected star-forming galaxies at similar redshifts. There are statistically significant anti-correlations found between the circularized half-light radius ($r_{\rm cir,50}$), the Ly$α$ equivalent width (EW(Ly$α$)), and the Ly$α$ escape faction ($f^{Lyα}_{esc}$), suggesting that small UV-continuum radii are crucial for Ly$α$ emission. GPs and high-redshift LAEs have similar sizes, once spatial resolution effects are properly considered. Our results show that a compact/small size is crucial for escape of Ly$α$ photons, and that Ly$α$-emitters show constant characteristic size independent of their redshift.

astro-ph.GA↗