arXiv subjects

Yang Fu

Publications and source records attributed to Yang Fu.

66 records · Page 4Linked to original sources

arXiv2020

CompFeat: Comprehensive Feature Aggregation for Video Instance Segmentation

Video instance segmentation is a complex task in which we need to detect, segment, and track each object for any given video. Previous approaches only utilize single-frame features for the detection, segmentation, and tracking of objects and they suffer in the video scenario due to several distinct challenges such as motion blur and drastic appearance change. To eliminate ambiguities introduced by only using single-frame features, we propose a novel comprehensive feature aggregation approach (CompFeat) to refine features at both frame-level and object-level with temporal and spatial context information. The aggregation process is carefully designed with a new attention mechanism which significantly increases the discriminative power of the learned features. We further improve the tracking capability of our model through a siamese design by incorporating both feature similarities and spatial similarities. Experiments conducted on the YouTube-VIS dataset validate the effectiveness of proposed CompFeat. Our code will be available at https://github.com/SHI-Labs/CompFeat-for-Video-Instance-Segmentation.

cs.CV↗

arXiv2020

Tunable layered-magnetism-assisted magneto-Raman effect in a two-dimensional magnet $\mathrm{CrI_3}$

We use a combination of polarized Raman spectroscopy experiment and model magnetism-phonon coupling calculations to study the rich magneto-Raman effect in the two-dimensional (2D) magnet $\mathrm{CrI_3}$. We reveal a novel layered-magnetism-assisted phonon scattering mechanism below the magnetic onset temperature, whose Raman excitation breaks time-reversal symmetry, has an antisymmetric Raman tensor, and follows the magnetic phase transitions across critical magnetic fields, on top of the presence of the conventional phonon scattering with symmetric Raman tensors in $N$-layer $\mathrm{CrI_3}$. We resolve in data and by calculations that the 1st-order $A_g$ phonon of monolayer splits into a $N$-fold multiplet in $N$-layer $\mathrm{CrI_3}$ due to the interlayer coupling ($N$>=2) and that the phonons with the multiple show distinct magnetic field dependence because of their different layered-magnetism-phonon coupling. We further find that such a layered-magnetism-phonon coupled Raman scattering mechanism extends beyond 1st-order to higher-order multi-phonon scattering processes. Our results on magneto-Raman effect of the 1st-order phonons in the multiplet and the higher-order multi-phonons in $N$-layer $\mathrm{CrI_3}$ demonstrate the rich and strong behavior of emergent magneto-optical effects in 2D magnets and underlines the unique opportunities of new spin-phonon physics in van der Waals layered magnets.

cond-mat.mtrl-sci↗

arXiv2020

Observation of the polaronic character of excitons in a two-dimensional semiconducting magnet $\mathrm{CrI_3}$

Exciton dynamics can be strongly affected by lattice vibrations through electron-phonon coupling. This is rarely explored in two-dimensional magnetic semiconductors. Focusing on bilayer CrI3, we first show the presence of strong electron-phonon coupling through temperature-dependent photoluminescence and absorption spectroscopy. We then report the observation of periodic broad modes up to the 8th order in Raman spectra, attributed to the polaronic character of excitons. We establish that this polaronic character is dominated by the coupling between the charge-transfer exciton at 1.96 eV and a longitudinal optical phonon at 120.6 cm-1. We further show that the emergence of long-range magnetic order enhances the electron-phonon coupling strength by about 50$\%$ and that the transition from layered antiferromagnetic to ferromagnetic order tunes the spectral intensity of the periodic broad modes, suggesting a strong coupling among the lattice, charge and spin in two-dimensional CrI3. Our study opens opportunities for tailoring light-matter interactions in two-dimensional magnetic semiconductors.

cond-mat.mes-hall↗

arXiv2020

Magnetic topological insulator MnBi6Te10 with zero-field ferromagnetic state and gapped Dirac surface states

Magnetic topological insulators (TIs) with nontrivial topological electronic structure and broken time-reversal symmetry exhibit various exotic topological quantum phenomena. The realization of such exotic phenomena at high temperature is one of central topics in this area. We reveal that MnBi6Te10 is a magnetic TI with an antiferromagnetic ground state below 10.8 K whose nontrivial topology is manifested by Dirac-like surface states. The ferromagnetic axion insulator state with Z4 = 2 emerges once spins polarized at field as low as 0.1 T, accompanied with saturated anomalous Hall resistivity up to 10 K. Such a ferromagnetic state is preserved even external field down to zero at 2 K. Theoretical calculations indicate that the few-layer ferromagnetic MnBi6Te10 is also topologically nontrivial with a non-zero Chern number. Angle-resolved photoemission spectroscopy experiments further reveal three types of Dirac surface states arising from different terminations on the cleavage surfaces, one of which has insulating behavior with an energy gap of ~ 28 meV at the Dirac point. These outstanding features suggest that MnBi6Te10 is a promising system to realize various topological quantum effects at zero field and high temperature.

cond-mat.mtrl-sci↗

arXiv2020

Lattice QCD calculation of the pion charge radius using a model-independent method

We use a method to calculate the hadron's charge radius without model-dependent momentum extrapolations. The method does not require the additional quark propagator inversions on the twisted boundary conditions or the computation of the momentum derivatives of quark propagators and thus is easy to implement. We apply this method to the calculation of pion charge radius $\langle r_π^2\rangle$. For comparison, we also determine $\langle r_π^2\rangle$ with the traditional approach of computing the slope of the form factors. The new method produces results consistent with those from the traditional method and with statistical errors 1.5-1.9 times smaller. For the four gauge ensembles at the physical pion masses, the statistical errors of $\langle r_π^2\rangle$ range from 2.1% to 4.6% by using $\lesssim50$ configurations. For the ensemble at $m_π\approx 340$ MeV, the statistical uncertainty is even reduced to a sub-percent level.

hep-lat↗

arXiv2019

Investigation upon the performance of piezoelectric energy harvester with elastic extensions

Piezoelectric vibration energy harvesters have attracted much attention due to its potential to replace currently popular batteries and to provide an sustainable power sources. Many researchers have proposed ways to increase the performance of piezoelectric energy harvesters like bandwidth, working frequency and output performance. Here in this contribution, we propose the method of using elastic extensions to tune the performance of a piezoelectric energy harvester. Mathematical model of the proposed device is derived and analyzed. Numerical simulations are done to investigate the influences of the derived parameters, like length ratio λ_l , bending stiffness ratio λ_B , and line density ratio λ_m . Results show that the elastic extension does change the motion of the proposed device and help tune the performance of piezoelectric energy harvesters.

physics.app-ph↗

arXiv2019

Self-similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-identification

Domain adaptation in person re-identification (re-ID) has always been a challenging task. In this work, we explore how to harness the natural similar characteristics existing in the samples from the target domain for learning to conduct person re-ID in an unsupervised manner. Concretely, we propose a Self-similarity Grouping (SSG) approach, which exploits the potential similarity (from global body to local parts) of unlabeled samples to automatically build multiple clusters from different views. These independent clusters are then assigned with labels, which serve as the pseudo identities to supervise the training process. We repeatedly and alternatively conduct such a grouping and training process until the model is stable. Despite the apparent simplify, our SSG outperforms the state-of-the-arts by more than 4.6% (DukeMTMC to Market1501) and 4.4% (Market1501 to DukeMTMC) in mAP, respectively. Upon our SSG, we further introduce a clustering-guided semisupervised approach named SSG ++ to conduct the one-shot domain adaption in an open set setting (i.e. the number of independent identities from the target domain is unknown). Without spending much effort on labeling, our SSG ++ can further promote the mAP upon SSG by 10.7% and 6.9%, respectively. Our Code is available at: https://github.com/OasisYang/SSG .

cs.CV↗

arXiv2018

Horizontal Pyramid Matching for Person Re-identification

Despite the remarkable recent progress, person re-identification (Re-ID) approaches are still suffering from the failure cases where the discriminative body parts are missing. To mitigate such cases, we propose a simple yet effective Horizontal Pyramid Matching (HPM) approach to fully exploit various partial information of a given person, so that correct person candidates can be still identified even even some key parts are missing. Within the HPM, we make the following contributions to produce a more robust feature representation for the Re-ID task: 1) we learn to classify using partial feature representations at different horizontal pyramid scales, which successfully enhance the discriminative capabilities of various person parts; 2) we exploit average and max pooling strategies to account for person-specific discriminative information in a global-local manner. To validate the effectiveness of the proposed HPM, extensive experiments are conducted on three popular benchmarks, including Market-1501, DukeMTMC-ReID and CUHK03. In particular, we achieve mAP scores of 83.1%, 74.5% and 59.7% on these benchmarks, which are the new state-of-the-arts. Our code is available on Github

cs.CV↗

arXiv2018

STA: Spatial-Temporal Attention for Large-Scale Video-based Person Re-Identification

In this work, we propose a novel Spatial-Temporal Attention (STA) approach to tackle the large-scale person re-identification task in videos. Different from the most existing methods, which simply compute representations of video clips using frame-level aggregation (e.g. average pooling), the proposed STA adopts a more effective way for producing robust clip-level feature representation. Concretely, our STA fully exploits those discriminative parts of one target person in both spatial and temporal dimensions, which results in a 2-D attention score matrix via inter-frame regularization to measure the importances of spatial parts across different frames. Thus, a more robust clip-level feature representation can be generated according to a weighted sum operation guided by the mined 2-D attention score matrix. In this way, the challenging cases for video-based person re-identification such as pose variation and partial occlusion can be well tackled by the STA. We conduct extensive experiments on two large-scale benchmarks, i.e. MARS and DukeMTMC-VideoReID. In particular, the mAP reaches 87.7% on MARS, which significantly outperforms the state-of-the-arts with a large margin of more than 11.6%.

cs.CV↗

arXiv2018

Observing of the super-Planckian near-field thermal radiation between graphene sheets

Thermal radiation can be substantially enhanced in the near-field scenario due to the tunneling of evanescent waves. The monolayer graphene could play a vital role in this process owning to its strong infrared plasmonic response, however, which still lacks an experimental verification due to the technical challenges. Here, we manage to make a direct measurement about plasmon-mediated thermal radiation between two macroscopic graphene sheets using a custom-made setup. Super-Planckian radiation with efficiency 4.5 times larger than the blackbody limit is observed at a 430-nm vacuum gap on insulating silicon hosting substrates. The positive role of graphene plasmons is further confirmed on conductive silicon substrates which have strong infrared loss and thermal emittance. Based on these, a thermophotovoltaic cell made of the graphene-silicon heterostructure is lastly discussed. The current work validates the classic thermodynamical theory in treating graphene and also paves a way to pursue the application of near-field thermal management.

physics.optics↗

arXiv2018

Daytime passive radiative cooler using porous alumina

We report a daytime passive radiative cooler using chemically fabricated porous anodic aluminum oxide (AAO) membranes. Effective medium theory (EMT) has been applied to analyzing the optical properties of the air-doped porous medium. The composite structure is specifically optimized so that it has a high absorbance (emittance) in the far-infrared atmospheric window and nearly no loss in the solar spectrum. The calculated emittance is well reproduced in the experiment by our AAO sample. The fabricated porous membrane shows a potential cooling power density of 64 W/m2 at ambient (humidity = 75%) under direct sunlight irradiance (AM1.5). Experimentally, the sample is cooled by a 2.6C temperature reduction below the ambient air temperature in the sunlight. This performance shows little variance at night. The AAO approach proposed here may provide a promising way to produce low-cost and efficient radiative cooler in large scales for feasible energy conservation.

physics.app-ph↗

arXiv2013

Photonic microwave generation with high-power photodiodes

We utilize and characterize high-power, high-linearity modified uni-traveling carrier (MUTC) photodiodes for low-phase-noise photonic microwave generation based on optical frequency division. When illuminated with picosecond pulses from a repetition-rate-multiplied gigahertz Ti:sapphire modelocked laser, the photodiodes can achieve 10 GHz signal power of +14 dBm. Using these diodes, a 10 GHz microwave tone is generated with less than 500 attoseconds absolute integrated timing jitter (1 Hz-10 MHz) and a phase noise floor of -177 dBc/Hz. We also characterize the electrical response, amplitude-to-phase conversion, saturation and residual noise of the MUTC photodiodes.

physics.optics↗