Searcharxiv⌕ Search

arXiv subjects

Zhi Li

Publications and source records attributed to Zhi Li.

At least 235 records · Page 13Linked to original sources
arXiv2022

Large-scale Hydrodynamical Shocks as the Smoking Gun Evidence for a Bar in M31

The formation and evolutionary history of M31 are closely related to its dynamical structures, which remain unclear due to its high inclination. Gas kinematics could provide crucial evidence for the existence of a rotating bar in M31. Using the position-velocity diagram of [OIII] and HI, we are able to identify clear sharp velocity jump (shock) features with a typical amplitude over 100 km/s in the central region of M31 (4.6 kpc X 2.3 kpc, or 20 arcmin X 10 arcmin). We also simulate gas morphology and kinematics in barred M31 potentials and find that the bar-induced shocks can produce velocity jumps similar to those in [OIII]. The identified shock features in both [OIII] and HI are broadly consistent, and they are found mainly on the leading sides of the bar/bulge, following a hallmark pattern expected from the bar-driven gas inflow. Shock features on the far side of the disk are clearer than those on the near side, possibly due to limited data coverage on the near side, as well as obscuration by the warped gas and dust layers. Further hydrodynamical simulations with more sophisticated physics are desired to fully understand the observed gas features and to better constrain the parameters of the bar in M31.

astro-ph.GA↗
arXiv2022

HULC: 3D Human Motion Capture with Pose Manifold Sampling and Dense Contact Guidance

Marker-less monocular 3D human motion capture (MoCap) with scene interactions is a challenging research topic relevant for extended reality, robotics and virtual avatar generation. Due to the inherent depth ambiguity of monocular settings, 3D motions captured with existing methods often contain severe artefacts such as incorrect body-scene inter-penetrations, jitter and body floating. To tackle these issues, we propose HULC, a new approach for 3D human MoCap which is aware of the scene geometry. HULC estimates 3D poses and dense body-environment surface contacts for improved 3D localisations, as well as the absolute scale of the subject. Furthermore, we introduce a 3D pose trajectory optimisation based on a novel pose manifold sampling that resolves erroneous body-environment inter-penetrations. Although the proposed method requires less structured inputs compared to existing scene-aware monocular MoCap algorithms, it produces more physically-plausible poses: HULC significantly and consistently outperforms the existing approaches in various experiments and on different metrics. Project page: https://vcai.mpi-inf.mpg.de/projects/HULC/.

cs.CV↗
arXiv2022

Weakly-Supervised Temporal Action Detection for Fine-Grained Videos with Hierarchical Atomic Actions

Action understanding has evolved into the era of fine granularity, as most human behaviors in real life have only minor differences. To detect these fine-grained actions accurately in a label-efficient way, we tackle the problem of weakly-supervised fine-grained temporal action detection in videos for the first time. Without the careful design to capture subtle differences between fine-grained actions, previous weakly-supervised models for general action detection cannot perform well in the fine-grained setting. We propose to model actions as the combinations of reusable atomic actions which are automatically discovered from data through self-supervised clustering, in order to capture the commonality and individuality of fine-grained actions. The learnt atomic actions, represented by visual concepts, are further mapped to fine and coarse action labels leveraging the semantic label hierarchy. Our approach constructs a visual representation hierarchy of four levels: clip level, atomic action level, fine action class level and coarse action class level, with supervision at each level. Extensive experiments on two large-scale fine-grained video datasets, FineAction and FineGym, show the benefit of our proposed weakly-supervised model for fine-grained action detection, and it achieves state-of-the-art results.

cs.CV↗
arXiv2022

Amorphous alloys surpass E/10 strength limit at extreme strain rates

Theoretical predictions of the ideal strength of materials range from E/30 to E/10 (E is Young's modulus). However, despite intense interest over the last decade, the value of the ideal strength that can be attained experimentally for metals remains a mystery (1-5). In this study, we demonstrated the unprecedented strength of an amorphous Cu-Zr alloy that surpassed the E/10 limit. Laser-induced shock experiments were conducted on Cu50Zr50 to explore its strength and failure mechanisms at ultrahigh strain rates. The material demonstrated a high spall strength of 9.8 GPa, approximately 1/13 of its P-wave modulus (~ E/6), at strain rates greater than 10^7 s^-1, which sets a new record for the elastic limit of metallic materials. Electron microscopy and large-scale molecular dynamics simulations revealed that void nucleation and growth, not shear-banding, comprised the major failure mechanism for metallic glasses at extremely fast strain rates. A new model for void formation under the control of surface energy explained the rate dependence of the material strength. The results of this study reveal new possible ways to use the amorphous phase in nanostructured metals in future applications under demanding mechanical conditions.

cond-mat.mtrl-sci↗
arXiv2022

PF4Microservices: A decomposion scheme for microservices based on Problem Frames

In recent years, microservice architecture has become a popular architectural style in software engineering, with its natural support for DevOps and continuous delivery, as well as its scalability and extensibility, which drive industry practitioners to migrate to microservice architecture. However, there are many challenges in adopting a microservice architecture, the most important of which is how to properly decomposition a monolithic system into microservices. Currently, microservice decomposition decisions for monolithic systems rely on subjective human experience, which is a costly, time-consuming process with high uncertainty of results. To address this problem, this paper proposes a method for microservice decomposition using Jackson Problem Frames. In this method, requirements of the system are analysed, descriptions of the interactions between the proposed software and its environment is obtained, multiple problem diagrams are constructed, and then the problem diagrams are merged by analyzing the correlation and similarity between them, resulting in a microservice decomposition scheme. A case study is also conducted based on a smart parking system. The results of the study show that the method can perform microservice decomposition based on requirements and the software environment, resulting in reducing the decisionmaking burden of developers, with reasonable decomposition results.

cs.SE↗
arXiv2022

FourNetFlows: An efficient model for steady airfoil flows prediction

FourNetFlows, the abbreviation of Fourier Neural Network for Airfoil Flows, is an efficient model that provides quick and accurate predictions of steady airfoil flows. We choose the Fourier Neural Operator (FNO) as the backbone architecture and utilize OpenFOAM to generate numerical solutions of airfoil flows for training. Our results indicate that FourNetFlows matches the accuracy of the Semi-Implicit Method for Pressure Linked Equations (SIMPLE) integrated with the Spalart-Allmaras turbulence model, one of the numerical algorithms. FourNetFlows is also used to predict flows around an oval whose shape is definitely different from samples in the training set. We note that both qualitative and quantitative results are consistent with the numerical results. Meanwhile, FourNetFlows solves thousands of solutions in seconds, orders of magnitude faster than the classical numerical method. Surprisingly, FourNetFlows achieves model flows with zero-shot super-resolution when it is trained under a lower resolution. And the inferring time is almost constant when the resolution of solutions is increasing.

physics.flu-dyn↗
arXiv2022

Topological classification of non-Hermitian bands

We proposed a framework for the topological classification of non-Hermitian systems. Different from previous $K$-theoretical approaches, our approach is a homotopy classification, which enables us to see more topological invariants. Specifically, we considered the classification of non-Hermitian systems with separable band structures. We found that the whole classification set is decomposed into several sectors based on the braiding of energy levels and characterized by some braid group data. Each sector can be further classified based on the topology of eigenstates (wave functions). Due to the interplay between energy levels braiding and eigenstates topology, we found some torsion invariants, which only appear in the non-Hermitian world via homotopical approach. We further proved that these new topological invariants are unstable (fragile), in the sense that adding more bands will trivialize these invariants.

cond-mat.mes-hall↗
arXiv2022

Emergent phase transition in Cluster Ising model with dissipation

We study a cluster Ising model with non-Hermitian external field which can be exactly solved in the language of free fermions. By investigating the second derivative of energy density and fidelity, the possible new critical points are tentatively located. String order parameter and staggered magnetization are then detected to reveal emergent phases of brand new characteristics. To categorize the exotic phases and phase transitions induced by non-Hermiticity, we calculate the variation mode of spin correlation function as well as string parameter, which characterize the emergent phases and critical points with different patterns of decay and critical exponents. With the help of string order parameter and staggered magnetization, we find that there are four phases after introducing the non-Hermiticity -- the cluster phase, the gapless phase, the paramagnetic (PM) phase and the antiferromagnetic (AF) phase. A phase diagram is then presented to graphically illustrate, based on two "KT-like" phase transitions and an Ising phase transition, respectively, the generation of three critical lines as non-Hermitian strength increases. Our theoretical work is expected to be realized in the experiment of ultra-cold atoms, pushing for progress in exploring novel phases and phase transitions.

cond-mat.stat-mech↗
arXiv2022

Learning Meta Pattern for Face Anti-Spoofing

Face Anti-Spoofing (FAS) is essential to secure face recognition systems and has been extensively studied in recent years. Although deep neural networks (DNNs) for the FAS task have achieved promising results in intra-dataset experiments with similar distributions of training and testing data, the DNNs' generalization ability is limited under the cross-domain scenarios with different distributions of training and testing data. To improve the generalization ability, recent hybrid methods have been explored to extract task-aware handcrafted features (e.g., Local Binary Pattern) as discriminative information for the input of DNNs. However, the handcrafted feature extraction relies on experts' domain knowledge, and how to choose appropriate handcrafted features is underexplored. To this end, we propose a learnable network to extract Meta Pattern (MP) in our learning-to-learn framework. By replacing handcrafted features with the MP, the discriminative information from MP is capable of learning a more generalized model. Moreover, we devise a two-stream network to hierarchically fuse the input RGB image and the extracted MP by using our proposed Hierarchical Fusion Module (HFM). We conduct comprehensive experiments and show that our MP outperforms the compared handcrafted features. Also, our proposed method with HFM and the MP can achieve state-of-the-art performance on two different domain generalization evaluation benchmarks.

cs.CV↗
arXiv2022

Short Proofs of Linear Growth of Quantum Circuit Complexity

The complexity of a quantum gate, defined as the minimal number of elementary gates to build it, is an important concept in quantum information and computation. It is shown recently that the complexity of quantum gates built from random quantum circuits almost surely grows linearly with the number of building blocks. In this article, we provide two short proofs of this fact. We also discuss a discrete version of quantum circuit complexity growth.

quant-ph↗
arXiv2022

One-Class Knowledge Distillation for Face Presentation Attack Detection

Face presentation attack detection (PAD) has been extensively studied by research communities to enhance the security of face recognition systems. Although existing methods have achieved good performance on testing data with similar distribution as the training data, their performance degrades severely in application scenarios with data of unseen distributions. In situations where the training and testing data are drawn from different domains, a typical approach is to apply domain adaptation techniques to improve face PAD performance with the help of target domain data. However, it has always been a non-trivial challenge to collect sufficient data samples in the target domain, especially for attack samples. This paper introduces a teacher-student framework to improve the cross-domain performance of face PAD with one-class domain adaptation. In addition to the source domain data, the framework utilizes only a few genuine face samples of the target domain. Under this framework, a teacher network is trained with source domain samples to provide discriminative feature representations for face PAD. Student networks are trained to mimic the teacher network and learn similar representations for genuine face samples of the target domain. In the test phase, the similarity score between the representations of the teacher and student networks is used to distinguish attacks from genuine ones. To evaluate the proposed framework under one-class domain adaptation settings, we devised two new protocols and conducted extensive experiments. The experimental results show that our method outperforms baselines under one-class domain adaptation settings and even state-of-the-art methods with unsupervised domain adaptation.

cs.CV↗
arXiv2022

Estimating the Resize Parameter in End-to-end Learned Image Compression

We describe a search-free resizing framework that can further improve the rate-distortion tradeoff of recent learned image compression models. Our approach is simple: compose a pair of differentiable downsampling/upsampling layers that sandwich a neural compression model. To determine resize factors for different inputs, we utilize another neural network jointly trained with the compression model, with the end goal of minimizing the rate-distortion objective. Our results suggest that "compression friendly" downsampled representations can be quickly determined during encoding by using an auxiliary network and differentiable image warping. By conducting extensive experimental tests on existing deep image compression models, we show results that our new resizing parameter estimation framework can provide Bjøntegaard-Delta rate (BD-rate) improvement of about 10% against leading perceptual quality engines. We also carried out a subjective quality study, the results of which show that our new approach yields favorable compressed images. To facilitate reproducible research in this direction, the implementation used in this paper is being made freely available online at: https://github.com/treammm/ResizeCompression.

eess.IV↗
arXiv2022

Preference Enhanced Social Influence Modeling for Network-Aware Cascade Prediction

Network-aware cascade size prediction aims to predict the final reposted number of user-generated information via modeling the propagation process in social networks. Estimating the user's reposting probability by social influence, namely state activation plays an important role in the information diffusion process. Therefore, Graph Neural Networks (GNN), which can simulate the information interaction between nodes, has been proved as an effective scheme to handle this prediction task. However, existing studies including GNN-based models usually neglect a vital factor of user's preference which influences the state activation deeply. To that end, we propose a novel framework to promote cascade size prediction by enhancing the user preference modeling according to three stages, i.e., preference topics generation, preference shift modeling, and social influence activation. Our end-to-end method makes the user activating process of information diffusion more adaptive and accurate. Extensive experiments on two large-scale real-world datasets have clearly demonstrated the effectiveness of our proposed model compared to state-of-the-art baselines.

cs.SI↗
arXiv2022

Dynamics in an exact solvable quantum magnet: benchmark for quantum computer

Quantum magnets are never short of novel and fascinating dynamics, yet its simulation by classical computers requires exponentially-scaled computation resources, which renders the research on large-scale many-body dynamics fiendishly difficult. In this letter, we explore the dynamic behavior of 2D large-scale ferromagnetic J1-J2 Heisenberg model both theoretically and experimentally. First, the analytical solution of magnon dynamics is obtained to show an obvious ballistic propagation of magnon, which is typical for quantum walk. Then, we verify the dynamic behavior of the system through numerical approach of exact diagonalization and tensor network method. We also calculate out-of-time ordered correlators and butterfly velocities among different lattice points, finding that they can well depict the competition between different couplings. Finally, a quantum walk experiment is designed and conducted on the basis of IBM programmable quantum processors, and the experimental results are in consistence with our theoretical predictions. Since the analytical results can be used, in principle, to predict the behavior of large-scale quantum many-body systems and even those infinitely large, this work will help facilitate further research on quantum walk and quantum many-body dynamics in large-scale lattice systems, guide future design of quantum computers, as well as popularize quantum computers until they are known and available to every household in the world.

quant-ph↗
arXiv2022

A Preliminary Research on Space Situational Awareness Based on Event Cameras

Event camera is a new type of sensor that is different from traditional cameras. Each pixel is triggered asynchronously by an event. The trigger event is the change of the brightness irradiated on the pixel. If the increment or decrement is higher than a certain threshold, the event is output. Compared with traditional cameras, event cameras have the advantages of high temporal resolution, low latency, high dynamic range, low bandwidth and low power consumption. We carried out a series of observation experiments in a simulated space lighting environment. The experimental results show that the event camera can give full play to the above advantages in space situational awareness. This article first introduces the basic principles of the event camera, then analyzes its advantages and disadvantages, then introduces the observation experiment and analyzes the experimental results, and finally, a workflow of space situational awareness based on event cameras is given.

cs.CV↗
arXiv2022

Improving Maximum Likelihood Difference Scaling method to measure inter content scale

The goal of most subjective studies is to place a set of stimuli on a perceptual scale. This is mostly done directly by rating, e.g. using single or double stimulus methodologies, or indirectly by ranking or pairwise comparison. All these methods estimate the perceptual magnitudes of the stimuli on a scale. However, procedures such as Maximum Likelihood Difference Scaling (MLDS) have shown that considering perceptual distances can bring benefits in terms of discriminatory power, observers' cognitive load, and the number of trials required. One of the disadvantages of the MLDS method is that the perceptual scales obtained for stimuli created from different source content are generally not comparable. In this paper, we propose an extension of the MLDS method that ensures inter-content comparability of the results and shows its usefulness especially in the presence of observer errors.

q-bio.NC↗
arXiv2022

Banding vs. Quality: Perceptual Impact and Objective Assessment

Staircase-like contours introduced to a video by quantization in flat areas, commonly known as banding, have been a long-standing problem in both video processing and quality assessment communities. The fact that even a relatively small change of the original pixel values can result in a strong impact on perceived quality makes banding especially difficult to be detected by objective quality metrics. In this paper, we study how banding annoyance compares to more commonly studied scaling and compression artifacts with respect to the overall perceptual quality. We further propose a simple combination of VMAF and the recently developed banding index, CAMBI, into a banding-aware video quality metric showing improved correlation with overall perceived quality.

cs.MM↗