arXiv subjects

Xu Wei

Publications and source records attributed to Xu Wei.

11 recordsLinked to original sources

arXiv2025

Long-Time Existence of Quasilinear Wave Equations Exterior to Star-shaped Obstacle in $2\mathbf{D}$

In this paper, we study the long-time existence result for small data solutions of quasilinear wave equations exterior to star-shaped regions in two space dimensions. The key novelty is that we establish a Morawetz type energy estimate for the perturbed inhomogeneous wave equation in the exterior domain, which yields $t^{-\frac12}$ decay inside the cone. In addition, two new weighted $L^2$ product estimates are established to produce $t^{-\frac12}$ decay close to the cone. We then show that the existence lifespan $T_\e$ for the quasilinear wave equations with general quadratic nonlinearity satisfies \begin{equation*} \varepsilon^2T_{\varepsilon}\ln^3T_{\varepsilon}=A, \end{equation*} for some fixed positive constant $A$, which is almost sharp (with some logarithmic loss) comparing to the known result of the corresponding Cauchy problem.

math.AP↗

arXiv2023

Blind Image Quality Assessment Using Multi-Stream Architecture with Spatial and Channel Attention

BIQA (Blind Image Quality Assessment) is an important field of study that evaluates images automatically. Although significant progress has been made, blind image quality assessment remains a difficult task since images vary in content and distortions. Most algorithms generate quality without emphasizing the important region of interest. In order to solve this, a multi-stream spatial and channel attention-based algorithm is being proposed. This algorithm generates more accurate predictions with a high correlation to human perceptual assessment by combining hybrid features from two different backbones, followed by spatial and channel attention to provide high weights to the region of interest. Four legacy image quality assessment datasets are used to validate the effectiveness of our proposed approach. Authentic and synthetic distortion image databases are used to demonstrate the effectiveness of the proposed method, and we show that it has excellent generalization properties with a particular focus on the perceptual foreground information.

cs.CV↗

arXiv2023

MedLSAM: Localize and Segment Anything Model for 3D CT Images

Recent advancements in foundation models have shown significant potential in medical image analysis. However, there is still a gap in models specifically designed for medical image localization. To address this, we introduce MedLAM, a 3D medical foundation localization model that accurately identifies any anatomical part within the body using only a few template scans. MedLAM employs two self-supervision tasks: unified anatomical mapping (UAM) and multi-scale similarity (MSS) across a comprehensive dataset of 14,012 CT scans. Furthermore, we developed MedLSAM by integrating MedLAM with the Segment Anything Model (SAM). This innovative framework requires extreme point annotations across three directions on several templates to enable MedLAM to locate the target anatomical structure in the image, with SAM performing the segmentation. It significantly reduces the amount of manual annotation required by SAM in 3D medical imaging scenarios. We conducted extensive experiments on two 3D datasets covering 38 distinct organs. Our findings are twofold: 1) MedLAM can directly localize anatomical structures using just a few template scans, achieving performance comparable to fully supervised models; 2) MedLSAM closely matches the performance of SAM and its specialized medical adaptations with manual prompts, while minimizing the need for extensive point annotations across the entire dataset. Moreover, MedLAM has the potential to be seamlessly integrated with future 3D SAM models, paving the way for enhanced segmentation performance. Our code is public at \href{https://github.com/openmedlab/MedLSAM}

cs.CV↗

arXiv2023

Organic molecules as origin of visible-range single photon emission from hexagonal boron nitride and mica

The discovery of room-temperature single-photon emitters (SPEs) hosted by two-dimensional hexagonal boron nitride (2D hBN) has sparked intense research interest. Although emitters in the vicinity of 2 eV have been studied extensively, their microscopic identity has remained elusive. The discussion of this class of SPEs has centered on point defects in the hBN crystal lattice, but none of the candidate defect structures have been able to capture the great heterogeneity in emitter properties that is observed experimentally. Employing a widely used sample preparation protocol but disentangling several confounding factors, we demonstrate conclusively that heterogeneous single-photon emission ~2 eV associated with hBN originates from organic molecules, presumably aromatic fluorophores. The appearance of those SPEs depends critically on the presence of organic processing residues during sample preparation, and emitters formed during heat treatment are not located within the hBN crystal as previously thought, but at the hBN/substrate interface. We further demonstrate that the same class of SPEs can be observed in a different 2D insulator, fluorophlogopite mica.

physics.optics↗

arXiv2020

Hybrid aerial ground locomotion with a single passive wheel

Exploiting contacts with environment structures provides extra force support to a UAV, often reducing the power consumption and hence extending the mission time. This paper investigates one such way to exploit flat surfaces in the environment by a novel aerial-ground hybrid locomotion. Our design is a single passive wheel integrated at the UAV bottom, serving a minimal design to date. We present the principle and implementation of such a simple design as well as its control. Flight experiments are conducted to verify the feasibility and the power saving caused by the ground locomotion. Results show that our minimal design allows successful aerial-ground hybrid locomotion even with a less-controllable bi-copter UAV. The ground locomotion saves up to 77% battery without much tuning effort.

cs.RO↗

arXiv2019

Deep Distance Transform for Tubular Structure Segmentation in CT Scans

Tubular structure segmentation in medical images, e.g., segmenting vessels in CT scans, serves as a vital step in the use of computers to aid in screening early stages of related diseases. But automatic tubular structure segmentation in CT scans is a challenging problem, due to issues such as poor contrast, noise and complicated background. A tubular structure usually has a cylinder-like shape which can be well represented by its skeleton and cross-sectional radii (scales). Inspired by this, we propose a geometry-aware tubular structure segmentation method, Deep Distance Transform (DDT), which combines intuitions from the classical distance transform for skeletonization and modern deep segmentation networks. DDT first learns a multi-task network to predict a segmentation mask for a tubular structure and a distance map. Each value in the map represents the distance from each tubular structure voxel to the tubular structure surface. Then the segmentation mask is refined by leveraging the shape prior reconstructed from the distance map. We apply our DDT on six medical image datasets. The experiments show that (1) DDT can boost tubular structure segmentation performance significantly (e.g., over 13% improvement measured by DSC for pancreatic duct segmentation), and (2) DDT additionally provides a geometrical measurement for a tubular structure, which is important for clinical diagnosis (e.g., the cross-sectional scale of a pancreatic duct can be an indicator for pancreatic cancer).

cs.CV↗

arXiv2018

An Advert Creation System for Next-Gen Publicity

With the rapid proliferation of multimedia data in the internet, there has been a fast rise in the creation of videos for the viewers. This enables the viewers to skip the advertisement breaks in the videos, using ad blockers and 'skip ad' buttons -- bringing online marketing and publicity to a stall. In this paper, we demonstrate a system that can effectively integrate a new advertisement into a video sequence. We use state-of-the-art techniques from deep learning and computational photogrammetry, for effective detection of existing adverts, and seamless integration of new adverts into video sequences. This is helpful for targeted advertisement, paving the path for next-gen publicity.

cs.MM↗

arXiv2017

Photoinduced coherent acoustic phonon dynamics inside Mott insulator Sr2IrO4 films observed by femtosecond X-ray pulses

We investigate the transient photoexcited lattice dynamics in a layered perovskite Mott insulator Sr2IrO4 by femtosecond X-ray diffraction using a laser plasma-based X-ray source. Ultrafast structural dynamics of Sr2IrO4 thin films are determined by observing the shift and broadening of the (0012) Bragg diffraction after excitation by 1.5 eV and 3.0 eV pump photons for films with different thicknesses. The observed transient lattice response can be well interpreted as a distinct three-step dynamics due to the propagation of coherent acoustic phonons generated by the photoinduced quasiparticles (QP). Employing a normalized phonon propagation model, we found that the photoinduced angular shifts of the Bragg peak collapse into a universal curve after introducing a normalized coordinates to account for different thicknesses and pump photon energies, pinpointing the origin of the lattice distortion and its early evolution. In addition, a transient photocurrent measurement indicates that the photoinduced QPs are charge neutral excitons. Mapping the phonon propagation and correlating its dynamics with the QP by ultrafast X-ray diffraction (UXRD) establish a powerful way to study electron-phonon coupling and uncover the exotic physics in strongly correlated systems under nonequilibrium conditions.

physics.atm-clus↗

arXiv2015

Pricing-based Distributed Energy-Efficient Beamforming for MISO Interference Channels

In this paper, we consider the problem of maximizing the weighted sum energy efficiency (WS-EE) for multi-input single-output (MISO) interference channels (ICs) which is well acknowledged as general models of heterogeneous networks (HetNets), multicell networks, etc. To address this problem, we develop an efficient distributed beamforming algorithm based on a pricing mechanism. Specifically, we carefully introduce a price metric for distributed beamforming design which fortunately allows efficient closed-form solutions to the per-user beam-vector optimization problem. The convergence of the distributed pricing-based beamforming design is theoretically proven. Furthermore, we present an implementation strategy of the proposed distributed algorithm with limited information exchange. Numerical results show that our algorithm converges much faster than existing algorithms, while yielding comparable, sometimes even better performance in terms of the WS-EE. Finally, by taking the backhaul power consumption into account, it is interesting to show that the proposed algorithm with limited information exchange achieves better WS-EE than the full information exchange based algorithm in some special cases.

cs.IT↗

arXiv2015

Large-scale Artificial Neural Network: MapReduce-based Deep Learning

Faced with continuously increasing scale of data, original back-propagation neural network based machine learning algorithm presents two non-trivial challenges: huge amount of data makes it difficult to maintain both efficiency and accuracy; redundant data aggravates the system workload. This project is mainly focused on the solution to the issues above, combining deep learning algorithm with cloud computing platform to deal with large-scale data. A MapReduce-based handwriting character recognizer will be designed in this project to verify the efficiency improvement this mechanism will achieve on training and practical large-scale data. Careful discussion and experiment will be developed to illustrate how deep learning algorithm works to train handwritten digits data, how MapReduce is implemented on deep learning neural network, and why this combination accelerates computation. Besides performance, the scalability and robustness will be mentioned in this report as well. Our system comes with two demonstration software that visually illustrates our handwritten digit recognition/encoding application.

cs.DC↗

arXiv2013

Conditioning of BPM pickup signals for operations of the Duke storage ring with a wide range of single-bunch current

The Duke storage ring is a dedicated driver for the storage ring based oscillator free-electron lasers (FELs), and the High Intensity Gamma-ray Source (HIGS). It is operated with a beam current ranging from about 1 mA to 100 mA per bunch for various operations and accelerator physics studies. High performance operations of the FEL and gamma-ray source require a stable electron beam orbit, which has been realized by the global orbit feedback system. As a critical part of the orbit feedback system, the electron beam position monitors (BPMs) are required to be able to precisely measure the electron beam orbit in a wide range of the single-bunch current. However, the high peak voltage of the BPM pickups associated with high single-bunch current degrades the performance of the BPM electronics, and can potentially damage the BPM electronics. A signal conditioning method using low pass filters is developed to reduce the peak voltage to protect the BPM electronics, and to make the BPMs capable of working with a wide range of single-bunch current. Simulations and electron beam based tests are performed. The results show that the Duke storage ring BPM system is capable of providing precise orbit measurements to ensure highly stable FEL and HIGS operations.

physics.acc-ph↗