BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Lei Sun

Lei Sun contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Computer VisionRoboticseess.IVastro-ph.HEMachine LearningArtificial IntelligenceCryptography and Securityphysics.opticsApplicationsastro-ph.GAcond-mat.mtrl-sciMethodologyNeural and Evolutionary Computingphysics.app-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
23works
0followers
14topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

23 published item(s)

preprint2026arXiv

DeformMaster: An Interactive Physics-Neural World Model for Deformable Objects from Videos

World models for deformable objects should recover not only geometry and appearance, but also underlying physical dynamics, interaction grounding, and material behavior. Learning such a model from real videos is challenging because deformable linear, planar, and volumetric objects evolve under high-dimensional deformation, noisy interactions, and complex material response. The model must therefore infer a physical state from visual observations, roll it forward under new interactions, and render the resulting dynamics with high visual fidelity. We present DeformMaster, a video-derived interactive physics--neural world model that turns real interaction videos into an online interactive model of deformable objects within a unified dynamics-and-appearance framework. DeformMaster preserves structured physical rollout while using a neural residual to compensate for unmodeled effects, grounds sparse hand motion as distributed compliant actuator for hand--continuum interaction, represents material response with spatially varying constitutive experts, and drives high-fidelity 4D appearance from the predicted physical evolution. Experiments on real-world deformable-object sequences demonstrate DeformMaster's ability to roll out future dynamics and render dynamic appearance, outperforming state-of-the-art baselines while supporting novel action rollout, material-parameter variation, and dynamic novel-view synthesis.

0 citations0 reviewsNo code linkedOpen access
preprint2025arXiv

Real-world Reinforcement Learning from Suboptimal Interventions

Real-world reinforcement learning (RL) offers a promising approach to training precise and dexterous robotic manipulation policies in an online manner, enabling robots to learn from their own experience while gradually reducing human labor. However, prior real-world RL methods often assume that human interventions are optimal across the entire state space, overlooking the fact that even expert operators cannot consistently provide optimal actions in all states or completely avoid mistakes. Indiscriminately mixing intervention data with robot-collected data inherits the sample inefficiency of RL, while purely imitating intervention data can ultimately degrade the final performance achievable by RL. The question of how to leverage potentially suboptimal and noisy human interventions to accelerate learning without being constrained by them thus remains open. To address this challenge, we propose SiLRI, a state-wise Lagrangian reinforcement learning algorithm for real-world robot manipulation tasks. Specifically, we formulate the online manipulation problem as a constrained RL optimization, where the constraint bound at each state is determined by the uncertainty of human interventions. We then introduce a state-wise Lagrange multiplier and solve the problem via a min-max optimization, jointly optimizing the policy and the Lagrange multiplier to reach a saddle point. Built upon a human-as-copilot teleoperation system, our algorithm is evaluated through real-world experiments on diverse manipulation tasks. Experimental results show that SiLRI effectively exploits human suboptimal interventions, reducing the time required to reach a 90% success rate by at least 50% compared with the state-of-the-art RL method HIL-SERL, and achieving a 100% success rate on long-horizon manipulation tasks where other RL methods struggle to succeed. Project website: https://silri-rl.github.io/.

0 citations0 reviewsNo code linkedOpen access
preprint2023arXiv

Event-Based Fusion for Motion Deblurring with Cross-modal Attention

Traditional frame-based cameras inevitably suffer from motion blur due to long exposure times. As a kind of bio-inspired camera, the event camera records the intensity changes in an asynchronous way with high temporal resolution, providing valid image degradation information within the exposure time. In this paper, we rethink the eventbased image deblurring problem and unfold it into an end-to-end two-stage image restoration network. To effectively fuse event and image features, we design an event-image cross-modal attention module applied at multiple levels of our network, which allows to focus on relevant features from the event branch and filter out noise. We also introduce a novel symmetric cumulative event representation specifically for image deblurring as well as an event mask gated connection between the two stages of our network which helps avoid information loss. At the dataset level, to foster event-based motion deblurring and to facilitate evaluation on challenging real-world images, we introduce the Real Event Blur (REBlur) dataset, captured with an event camera in an illumination controlled optical laboratory. Our Event Fusion Network (EFNet) sets the new state of the art in motion deblurring, surpassing both the prior best-performing image-based method and all event-based methods with public implementations on the GoPro dataset (by up to 2.47dB) and on our REBlur dataset, even in extreme blurry conditions. The code and our REBlur dataset will be made publicly available.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Additional evidence for a pulsar wind nebula in the heart of SN 1987A from multi-epoch X-ray data and MHD modeling

Since the day of its explosion, supernova (SN) 1987A has been closely monitored to study its evolution and to detect its central compact relic. In fact, the formation of a neutron star is strongly supported by the detection of neutrinos from the SN. However, besides the detection in the Atacama Large Millimeter/submillimeter Array (ALMA) data of a feature that is compatible with the emission arising from a proto-pulsar wind nebula (PWN), the only hint for the existence of such elusive compact object is provided by the detection of hard emission in NuSTAR data up to ~ 20 keV. We report on the simultaneous analysis of multi-epoch observations of SN 1987A performed with Chandra, XMM-Newton and NuSTAR. We also compare the observations with a state-of-the-art 3D magnetohydrodynamic (MHD) simulation of SN 1987A. A heavily absorbed power-law, consistent with the emission from a PWN embedded in the heart of SN 1987A, is needed to properly describe the high-energy part of the observed spectra. The spectral parameters of the best-fit power-law are in agreement with the previous estimate, and exclude diffusive shock acceleration as a possible mechanism responsible for the observed non-thermal emission. The information extracted from our analysis are used to infer the physical characteristics of the pulsar and the broad-band emission of its nebula, in agreement with the ALMA data. Analysis of the synthetic spectra also show that, in the near future, the main contribution to Fe K emission line will originate in the outermost shocked ejecta of SN 1987A.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Annular Computational Imaging: Capture Clear Panoramic Images through Simple Lens

Panoramic Annular Lens (PAL) composed of few lenses has great potential in panoramic surrounding sensing tasks for mobile and wearable devices because of its tiny size and large Field of View (FoV). However, the image quality of tiny-volume PAL confines to optical limit due to the lack of lenses for aberration correction. In this paper, we propose an Annular Computational Imaging (ACI) framework to break the optical limit of light-weight PAL design. To facilitate learning-based image restoration, we introduce a wave-based simulation pipeline for panoramic imaging and tackle the synthetic-to-real gap through multiple data distributions. The proposed pipeline can be easily adapted to any PAL with design parameters and is suitable for loose-tolerance designs. Furthermore, we design the Physics Informed Image Restoration Network (PI2RNet) considering the physical priors of panoramic imaging and single-pass physics-informed engine. At the dataset level, we create the DIVPano dataset and the extensive experiments on it illustrate that our proposed network sets the new state of the art in the panoramic image restoration under spatially-variant degradation. In addition, the evaluation of the proposed ACI on a simple PAL with only 3 spherical lenses reveals the delicate balance between high-quality panoramic imaging and compact design. To the best of our knowledge, we are the first to explore Computational Imaging (CI) in PAL. Code and datasets are publicly available at https://github.com/zju-jiangqi/ACI-PI2RNet.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

CE-based white-box adversarial attacks will not work using super-fitting

Deep neural networks are widely used in various fields because of their powerful performance. However, recent studies have shown that deep learning models are vulnerable to adversarial attacks, i.e., adding a slight perturbation to the input will make the model obtain wrong results. This is especially dangerous for some systems with high-security requirements, so this paper proposes a new defense method by using the model super-fitting state to improve the model's adversarial robustness (i.e., the accuracy under adversarial attacks). This paper mathematically proves the effectiveness of super-fitting and enables the model to reach this state quickly by minimizing unrelated category scores (MUCS). Theoretically, super-fitting can resist any existing (even future) CE-based white-box adversarial attacks. In addition, this paper uses a variety of powerful attack algorithms to evaluate the adversarial robustness of super-fitting, and the proposed method is compared with nearly 50 defense models from recent conferences. The experimental results show that the super-fitting method in this paper can make the trained model obtain the highest adversarial robustness.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Efficient Human Pose Estimation via 3D Event Point Cloud

Human Pose Estimation (HPE) based on RGB images has experienced a rapid development benefiting from deep learning. However, event-based HPE has not been fully studied, which remains great potential for applications in extreme scenes and efficiency-critical conditions. In this paper, we are the first to estimate 2D human pose directly from 3D event point cloud. We propose a novel representation of events, the rasterized event point cloud, aggregating events on the same position of a small time slice. It maintains the 3D features from multiple statistical cues and significantly reduces memory consumption and computation complexity, proved to be efficient in our work. We then leverage the rasterized event point cloud as input to three different backbones, PointNet, DGCNN, and Point Transformer, with two linear layer decoders to predict the location of human keypoints. We find that based on our method, PointNet achieves promising results with much faster speed, whereas Point Transfomer reaches much higher accuracy, even close to previous event-frame-based methods. A comprehensive set of results demonstrates that our proposed method is consistently effective for these 3D backbone models in event-driven human pose estimation. Our method based on PointNet with 2048 points input achieves 82.46mm in MPJPE3D on the DHP19 dataset, while only has a latency of 12.29ms on an NVIDIA Jetson Xavier NX edge computing platform, which is ideally suitable for real-time detection with event cameras. Code is available at https://github.com/MasterHow/EventPointPose.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

FaceFormer: Scale-aware Blind Face Restoration with Transformers

Blind face restoration usually encounters with diverse scale face inputs, especially in the real world. However, most of the current works support specific scale faces, which limits its application ability in real-world scenarios. In this work, we propose a novel scale-aware blind face restoration framework, named FaceFormer, which formulates facial feature restoration as scale-aware transformation. The proposed Facial Feature Up-sampling (FFUP) module dynamically generates upsampling filters based on the original scale-factor priors, which facilitate our network to adapt to arbitrary face scales. Moreover, we further propose the facial feature embedding (FFE) module which leverages transformer to hierarchically extract diversity and robustness of facial latent. Thus, our FaceFormer achieves fidelity and robustness restored faces, which possess realistic and symmetrical details of facial components. Extensive experiments demonstrate that our proposed method trained with synthetic dataset generalizes better to a natural low quality images than current state-of-the-arts.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Five-channel frequency-division multiplexing using low-loss epsilon-near-zero metamaterial waveguide

The rapidly growing global data usage has demanded more efficient ways to utilize the scarce electromagnetic spectrum resource. Recent research has focused on the development of efficient multiplexing techniques in the millimeter-wave band (1-10 mm, or 30-300 GHz) due to the promise of large available bandwidth for future wireless networks. Frequency-division multiplexing is still one of the most commonly-used techniques to maximize the transmission capacity of a wireless network. Based on the frequency-selective tunnelling effect of the low-loss epsilon-near-zero metamaterial waveguide, we numerically and experimentally demonstrate five-channel frequency-division multiplexing and demultiplexing in the millimeter-wave range. We show that this device architecture offers great flexibility to manipulate the filter Q-factors and the transmission spectra of different channels, by changing of the epsilon-near-zero metamaterial waveguide topology and by adding a standard waveguide between two epsilon-near-zero channels. This strategy of frequency-division multiplexing may pave a way for efficiently allocating the spectrum for future communication networks.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Measuring the severity of multi-collinearity in high dimensions

Multi-collinearity is a wide-spread phenomenon in modern statistical applications and when ignored, can negatively impact model selection and statistical inference. Classic tools and measures that were developed for "$n>p$" data are not applicable nor interpretable in the high-dimensional regime. Here we propose 1) new individualized measures that can be used to visualize patterns of multi-collinearity, and subsequently 2) global measures to assess the overall burden of multi-collinearity without limiting the observed data dimensions. We applied these measures to genomic applications to investigate patterns of multi-collinearity in genetic variations across individuals with diverse ancestral backgrounds. The measures were able to visually distinguish genomic regions of excessive multi-collinearity and contrast the level of multi-collinearity between different continental populations.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Multi-Task Learning Framework for Emotion Recognition in-the-wild

This paper presents our system for the Multi-Task Learning (MTL) Challenge in the 4th Affective Behavior Analysis in-the-wild (ABAW) competition. We explore the research problems of this challenge from three aspects: 1) For obtaining efficient and robust visual feature representations, we propose MAE-based unsupervised representation learning and IResNet/DenseNet-based supervised representation learning methods; 2) Considering the importance of temporal information in videos, we explore three types of sequential encoders to capture the temporal information, including the encoder based on transformer, the encoder based on LSTM, and the encoder based on GRU; 3) For modeling the correlation between these different tasks (i.e., valence, arousal, expression, and AU) for multi-task affective analysis, we first explore the dependency between these different tasks and propose three multi-task learning frameworks to model the correlations effectively. Our system achieves the performance of $1.7607$ on the validation dataset and $1.4361$ on the test dataset, ranking first in the MTL Challenge. The code is available at https://github.com/AIM3-RUC/ABAW4.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Novel boron nitride polymorphs with graphite-diamond hybrid structure

Both boron nitride (BN) and carbon (C) have sp, sp2 and sp3 hybridization modes, and thus resulting in a variety of BN and C polymorphs with similar structures, such as hexagonal BN (hBN) and graphite, cubic BN (cBN) and diamond. Here, five types of BN polymorph structures were proposed theoretically, inspired by the graphite-diamond hybrid structures discovered in recent experiment. These BN polymorphs with graphite-diamond hybrid structures possessed excellent mechanical properties with combined high hardness and high ductility, and also exhibited various electronic properties such as semi-conductivity, semi-metallicity, and even one- and two-dimensional conductivity, differing from known insulators hBN and cBN. The simulated diffraction patterns of these BN hybrid structures could account for the unsolved diffraction patterns of intermediate products composed of "compressed hBN" and diamond-like BN, caused by phase transitions in previous experiments. Thus, this work provides a theoretical basis for the presence of these types of hybrid materials during phase transitions between graphite-like and diamond-like BN polymorphs.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Results

This paper reviews the NTIRE 2022 challenge on efficient single image super-resolution with focus on the proposed solutions and results. The task of the challenge was to super-resolve an input image with a magnification factor of $\times$4 based on pairs of low and corresponding high resolution images. The aim was to design a network for single image super-resolution that achieved improvement of efficiency measured according to several metrics including runtime, parameters, FLOPs, activations, and memory consumption while at least maintaining the PSNR of 29.00dB on DIV2K validation set. IMDN is set as the baseline for efficiency measurement. The challenge had 3 tracks including the main track (runtime), sub-track one (model complexity), and sub-track two (overall performance). In the main track, the practical runtime performance of the submissions was evaluated. The rank of the teams were determined directly by the absolute value of the average runtime on the validation set and test set. In sub-track one, the number of parameters and FLOPs were considered. And the individual rankings of the two metrics were summed up to determine a final ranking in this track. In sub-track two, all of the five metrics mentioned in the description of the challenge including runtime, parameter count, FLOPs, activations, and memory consumption were considered. Similar to sub-track one, the rankings of five metrics were summed up to determine a final ranking. The challenge had 303 registered participants, and 43 teams made valid submissions. They gauge the state-of-the-art in efficient single image super-resolution.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

On the Connection between Local Attention and Dynamic Depth-wise Convolution

Vision Transformer (ViT) attains state-of-the-art performance in visual recognition, and the variant, Local Vision Transformer, makes further improvements. The major component in Local Vision Transformer, local attention, performs the attention separately over small local windows. We rephrase local attention as a channel-wise locally-connected layer and analyze it from two network regularization manners, sparse connectivity and weight sharing, as well as weight computation. Sparse connectivity: there is no connection across channels, and each position is connected to the positions within a small local window. Weight sharing: the connection weights for one position are shared across channels or within each group of channels. Dynamic weight: the connection weights are dynamically predicted according to each image instance. We point out that local attention resembles depth-wise convolution and its dynamic version in sparse connectivity. The main difference lies in weight sharing - depth-wise convolution shares connection weights (kernel weights) across spatial positions. We empirically observe that the models based on depth-wise convolution and the dynamic variant with lower computation complexity perform on-par with or sometimes slightly better than Swin Transformer, an instance of Local Vision Transformer, for ImageNet classification, COCO object detection and ADE semantic segmentation. These observations suggest that Local Vision Transformer takes advantage of two regularization forms and dynamic weight to increase the network capacity. Code is available at https://github.com/Atten4Vis/DemystifyLocalViT.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Real Image Restoration via Structure-preserving Complementarity Attention

Since convolutional neural networks perform well in learning generalizable image priors from large-scale data, these models have been widely used in image denoising tasks. However, the computational complexity increases dramatically as well on complex model. In this paper, We propose a novel lightweight Complementary Attention Module, which includes a density module and a sparse module, which can cooperatively mine dense and sparse features for feature complementary learning to build an efficient lightweight architecture. Moreover, to reduce the loss of details caused by denoising, this paper constructs a gradient-based structure-preserving branch. We utilize gradient-based branches to obtain additional structural priors for denoising, and make the model pay more attention to image geometric details through gradient loss optimization.Based on the above, we propose an efficiently Unet structured network with dual branch, the visual results show that can effectively preserve the structural details of the original image, we evaluate benchmarks including SIDD and DND, where SCANet achieves state-of-the-art performance in PSNR and SSIM while significantly reducing computational cost.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Rethinking Classifier and Adversarial Attack

Various defense models have been proposed to resist adversarial attack algorithms, but existing adversarial robustness evaluation methods always overestimate the adversarial robustness of these models (i.e., not approaching the lower bound of robustness). To solve this problem, this paper uses the proposed decouple space method to divide the classifier into two parts: non-linear and linear. Then, this paper defines the representation vector of the original example (and its space, i.e., the representation space) and uses the iterative optimization of Absolute Classification Boundaries Initialization (ACBI) to obtain a better attack starting point. Particularly, this paper applies ACBI to nearly 50 widely-used defense models (including 8 architectures). Experimental results show that ACBI achieves lower robust accuracy in all cases.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

TSRFormer: Table Structure Recognition with Transformers

We present a new table structure recognition (TSR) approach, called TSRFormer, to robustly recognizing the structures of complex tables with geometrical distortions from various table images. Unlike previous methods, we formulate table separation line prediction as a line regression problem instead of an image segmentation problem and propose a new two-stage DETR based separator prediction approach, dubbed \textbf{Sep}arator \textbf{RE}gression \textbf{TR}ansformer (SepRETR), to predict separation lines from table images directly. To make the two-stage DETR framework work efficiently and effectively for the separation line prediction task, we propose two improvements: 1) A prior-enhanced matching strategy to solve the slow convergence issue of DETR; 2) A new cross attention module to sample features from a high-resolution convolutional feature map directly so that high localization accuracy is achieved with low computational cost. After separation line prediction, a simple relation network based cell merging module is used to recover spanning cells. With these new techniques, our TSRFormer achieves state-of-the-art performance on several benchmark datasets, including SciTSR, PubTabNet and WTW. Furthermore, we have validated the robustness of our approach to tables with complex structures, borderless cells, large blank spaces, empty or spanning cells as well as distorted or even curved shapes on a more challenging real-world in-house dataset.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Unusually high HCO+/CO ratios in and outside supernova remnant W49B

Galactic supernova remnants (SNRs) and their environments provide the nearest laboratories to study SN feedback. We performed molecular observations toward SNR W49B, the most luminous Galactic SNR in the X-ray band, aiming to explore signs of multiple feedback channels of SNRs on nearby molecular clouds (MCs). We found very broad HCO+ lines with widths of dv = 48--75 km/s in the SNR southwest, providing strong evidence that W49B is perturbing MCs at a systemic velocity of $V_{LSR}=61$--65 km/s, and placing W49B at a distance of $7.9\pm 0.6$ kpc. We observed unusually high-intensity ratios of HCO+ J=1-0/CO J=1-0 not only at shocked regions ($1.1\pm 0.4$ and $0.70\pm 0.16$), but also in quiescent clouds over 1 pc away from the SNR's eastern boundary (> 0.2). By comparing with the magnetohydrodynamics shock models, we interpret that the high ratio in the broad-line regions can result from a cosmic-ray (CR) induced chemistry in shocked MCs, where the CR ionization rate is enhanced to around 10--100 times of the Galactic level. The high HCO+/CO ratio outside the SNR is probably caused by the radiation precursor, while the luminous X-ray emission of W49B can explain a few properties in this region. The above results provide observational evidence that SNRs can strongly influence the molecular chemistry in and outside the shock boundary via their shocks, CRs, and radiation. We propose that the HCO+/CO ratio is a potentially useful tool to probe an SNR's multichannel influence on MCs.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

TriVoC: Efficient Voting-based Consensus Maximization for Robust Point Cloud Registration with Extreme Outlier Ratios

Correspondence-based point cloud registration is a cornerstone in robotics perception and computer vision, which seeks to estimate the best rigid transformation aligning two point clouds from the putative correspondences. However, due to the limited robustness of 3D keypoint matching approaches, outliers, probably in large numbers, are prone to exist among the correspondences, which makes robust registration methods imperative. Unfortunately, existing robust methods have their own limitations (e.g. high computational cost or limited robustness) when facing high or extreme outlier ratios, probably unsuitable for practical use. In this paper, we present a novel, fast, deterministic and guaranteed robust solver, named TriVoC (Triple-layered Voting with Consensus maximization), for the robust registration problem. We decompose the selecting of the minimal 3-point sets into 3 consecutive layers, and in each layer we design an efficient voting and correspondence sorting framework on the basis of the pairwise equal-length constraint. In this manner, the 3-point sets can be selected independently from the reduced correspondence sets according to the sorted sequence, which can significantly lower the computational cost and meanwhile provide a strong guarantee to achieve the largest consensus set (as the final inlier set) as long as a probabilistic termination condition is fulfilled. Varied experiments show that our solver TriVoC is robust against up to 99% outliers, highly accurate, time-efficient even with extreme outlier ratios, and also practical for real-world applications, showing performance superior to other state-of-the-art competitors.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Adaptive Operator Selection Based on Dynamic Thompson Sampling for MOEA/D

In evolutionary computation, different reproduction operators have various search dynamics. To strike a well balance between exploration and exploitation, it is attractive to have an adaptive operator selection (AOS) mechanism that automatically chooses the most appropriate operator on the fly according to the current status. This paper proposes a new AOS mechanism for multi-objective evolutionary algorithm based on decomposition (MOEA/D). More specifically, the AOS is formulated as a multi-armed bandit problem where the dynamic Thompson sampling (DYTS) is applied to adapt the bandit learning model, originally proposed with an assumption of a fixed award distribution, to a non-stationary setup. In particular, each arm of our bandit learning model represents a reproduction operator and is assigned with a prior reward distribution. The parameters of these reward distributions will be progressively updated according to the performance of its performance collected from the evolutionary process. When generating an offspring, an operator is chosen by sampling from those reward distribution according to the DYTS. Experimental results fully demonstrate the effectiveness and competitiveness of our proposed AOS mechanism compared with other four state-of-the-art MOEA/D variants.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

An XMM-Newton X-ray View of Supernova Remnant W49B: Revisiting its Recombining Plasmas and Progenitor Type

We present a comprehensive X-ray spectroscopy and imaging study of supernova remnant W49B using archival XMM-Newton observations. The overionization state of the shocked ejecta in W49B is clearly indicated by the radiative recombination continua of Si XIV, S XV, and Fe XXV, combined with the Ly$α$ lines of Ca and Fe. The line flux images of W49B indicate high emission measures of the central bar-like region for almost all the emission lines, while the equivalent width maps reveal a stratified structure for the metal abundance distributions. The global spectrum of W49B is well reproduced by a model containing one collisional ionization equilibrium (CIE) plasma component and two recombining plasma (RP) components. The CIE plasma represents the shocked interstellar medium, which dominates the X-ray emitting volume in W49B with a mass $\sim450M_\odot$. The two RP components with a total mass $\sim4.6M_\odot$ are both dominated by the ejecta material, but characterized by different electron temperatures ($\sim1.60$ keV and $\sim0.64$ keV). The recombination ages of the RP components are estimated as $\sim6000$ yr and $\sim3400$ yr, respectively. We then reveal the possibility of a thermal conduction origin for the high-temperature RP in W49B by calculating the conduction timescale. The metal abundance ratios of the ejecta in W49B are roughly consistent with a core-collapse explosion model with a $\lesssim15M_\odot$ progenitor, except for a rather high Mn/Fe. A Type Ia origin can explain the Mn abundance, while it predicts much higher ejecta masses than observed values for all the metal species considered in our analysis.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Real-time Fusion Network for RGB-D Semantic Segmentation Incorporating Unexpected Obstacle Detection for Road-driving Images

Semantic segmentation has made striking progress due to the success of deep convolutional neural networks. Considering the demands of autonomous driving, real-time semantic segmentation has become a research hotspot these years. However, few real-time RGB-D fusion semantic segmentation studies are carried out despite readily accessible depth information nowadays. In this paper, we propose a real-time fusion semantic segmentation network termed RFNet that effectively exploits complementary cross-modal information. Building on an efficient network architecture, RFNet is capable of running swiftly, which satisfies autonomous vehicles applications. Multi-dataset training is leveraged to incorporate unexpected small obstacle detection, enriching the recognizable classes required to face unforeseen hazards in the real world. A comprehensive set of experiments demonstrates the effectiveness of our framework. On Cityscapes, Our method outperforms previous state-of-the-art semantic segmenters, with excellent accuracy and 22Hz inference speed at the full 2048x1024 resolution, outperforming most existing RGB-D networks.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

ReLaText: Exploiting Visual Relationships for Arbitrary-Shaped Scene Text Detection with Graph Convolutional Networks

We introduce a new arbitrary-shaped text detection approach named ReLaText by formulating text detection as a visual relationship detection problem. To demonstrate the effectiveness of this new formulation, we start from using a "link" relationship to address the challenging text-line grouping problem firstly. The key idea is to decompose text detection into two subproblems, namely detection of text primitives and prediction of link relationships between nearby text primitive pairs. Specifically, an anchor-free region proposal network based text detector is first used to detect text primitives of different scales from different feature maps of a feature pyramid network, from which a text primitive graph is constructed by linking each pair of nearby text primitives detected from a same feature map with an edge. Then, a Graph Convolutional Network (GCN) based link relationship prediction module is used to prune wrongly-linked edges in the text primitive graph to generate a number of disjoint subgraphs, each representing a detected text instance. As GCN can effectively leverage context information to improve link prediction accuracy, our GCN based text-line grouping approach can achieve better text detection accuracy than previous text-line grouping methods, especially when dealing with text instances with large inter-character or very small inter-line spacings. Consequently, the proposed ReLaText achieves state-of-the-art performance on five public text detection benchmarks, namely RCTW-17, MSRA-TD500, Total-Text, CTW1500 and DAST1500.

0 citations0 reviewsNo code linkedOpen access