arXiv subjects

Lei Qiao

Publications and source records attributed to Lei Qiao.

At least 37 records · Page 2Linked to original sources

arXiv2022

Backbone is All Your Need: A Simplified Architecture for Visual Object Tracking

Exploiting a general-purpose neural architecture to replace hand-wired designs or inductive biases has recently drawn extensive interest. However, existing tracking approaches rely on customized sub-modules and need prior knowledge for architecture selection, hindering the tracking development in a more general system. This paper presents a Simplified Tracking architecture (SimTrack) by leveraging a transformer backbone for joint feature extraction and interaction. Unlike existing Siamese trackers, we serialize the input images and concatenate them directly before the one-branch backbone. Feature interaction in the backbone helps to remove well-designed interaction modules and produce a more efficient and effective framework. To reduce the information loss from down-sampling in vision transformers, we further propose a foveal window strategy, providing more diverse input patches with acceptable computational costs. Our SimTrack improves the baseline with 2.5%/2.6% AUC gains on LaSOT/TNL2K and gets results competitive with other specialized tracking algorithms without bells and whistles.

cs.CV↗

arXiv2022

Unsupervised Learning of Accurate Siamese Tracking

Unsupervised learning has been popular in various computer vision tasks, including visual object tracking. However, prior unsupervised tracking approaches rely heavily on spatial supervision from template-search pairs and are still unable to track objects with strong variation over a long time span. As unlimited self-supervision signals can be obtained by tracking a video along a cycle in time, we investigate evolving a Siamese tracker by tracking videos forward-backward. We present a novel unsupervised tracking framework, in which we can learn temporal correspondence both on the classification branch and regression branch. Specifically, to propagate reliable template feature in the forward propagation process so that the tracker can be trained in the cycle, we first propose a consistency propagation transformation. We then identify an ill-posed penalty problem in conventional cycle training in backward propagation process. Thus, a differentiable region mask is proposed to select features as well as to implicitly penalize tracking errors on intermediate frames. Moreover, since noisy labels may degrade training, we propose a mask-guided loss reweighting strategy to assign dynamic weights based on the quality of pseudo labels. In extensive experiments, our tracker outperforms preceding unsupervised methods by a substantial margin, performing on par with supervised methods on large-scale datasets such as TrackingNet and LaSOT. Code is available at https://github.com/FlorinShum/ULAST.

cs.CV↗

arXiv2022

Coherent Control of Collective Spontaneous Emission through Self-interference

As one of the central topics in quantum optics, collective spontaneous emission such as superradiance has been realized in a variety of systems. This work proposes an innovative scheme to coherently control collective emission rates via a self-interference mechanism in a nonlinear waveguide setting. The self-interference is made possible by photon backward scattering incurred by quantum scatterers in a waveguide working as quantum switches. Whether the interference is constructive or destructive is found to depend strongly on the distance between the scatterers and the emitters. The interference between two propagation pathways of the same photon leads to controllable superradiance and subradiance, with their collective decay rates much enhanced or suppressed (also leading to hyperradiance or population trapping). Furthermore, the self-interference mechanism is manifested by an abrupt change in the emission rates in real time. An experimental setup based on superconducting transmission line resonators and transmon qubits is further proposed to realize controllable collective emission rates.

quant-ph↗

arXiv2021

Coexistence of superconductivity and antiferromagentic order in Er$_{2}$O$_{2}$Bi with anti-ThCr$_{2}$Si$_{2}$ structure

We investigated the coexistence of superconductivity and antiferromagnetic order in the compound Er$_{2}$O$_{2}$Bi with anti-ThCr$_{2}$Si$_{2}$-type structure through resistivity, magnetization, specific heat measurements and first-principle calculations. The superconducting transition temperature $T_{\rm c}$ of 1.23 K and antiferromagnetic transition temperature $T_{\rm N}$ of 3 K are observed in the sample with the best nominal composition. The superconducting upper critical field $H_{\rm c2}$(0) and electron-phonon coupling constant $λ$$_{e-ph}$ in Er$_{2}$O$_{2}$Bi are similar to those in the previously reported non-magnetic superconductor Y$_{2}$O$_{2}$Bi with the same structure, indicating that the superconductivity in Er$_{2}$O$_{2}$Bi may have the same origin as in Y$_{2}$O$_{2}$Bi. The first-principle calculations of Er$_{2}$O$_{2}$Bi show that the Fermi surface is mainly composed of the Bi 6$p$ orbitals both in the paramagnetic and antiferromagnetic state, implying minor effect of the 4$f$ electrons on the Fermi surface. Besides, upon increasing the oxygen incorporation in Er$_{2}$O$_{x}$Bi, $T_{\rm c}$ increases from 1 to 1.23 K and $T_{\rm N}$ decreases slightly from 3 to 2.96 K, revealing that superconductivity and antiferromagnetic order may compete with each other. The Hall effect measurements indicate that hole-type carrier density indeed increases with increasing oxygen content, which may account for the variations of $T_{\rm c}$ and $T_{\rm N}$ with different oxygen content.

cond-mat.supr-con↗

arXiv2021

Robust perfect equilibrium in large games

This paper proposes a new equilibrium concept "robust perfect equilibrium" for non-cooperative games with a continuum of players, incorporating three types of perturbations. Such an equilibrium is shown to exist (in symmetric mixed strategies and in pure strategies) and satisfy the important properties of admissibility, aggregate robustness, and ex post robust perfection. These properties strengthen relevant equilibrium results in an extensive literature on strategic interactions among a large number of agents. Illustrative applications to congestion games and potential games are presented. In the particular case of a congestion game with strictly increasing cost functions, we show that there is a unique symmetric robust perfect equilibrium.

econ.TH↗

arXiv2021

Higher Performance Visual Tracking with Dual-Modal Localization

Visual Object Tracking (VOT) has synchronous needs for both robustness and accuracy. While most existing works fail to operate simultaneously on both, we investigate in this work the problem of conflicting performance between accuracy and robustness. We first conduct a systematic comparison among existing methods and analyze their restrictions in terms of accuracy and robustness. Specifically, 4 formulations-offline classification (OFC), offline regression (OFR), online classification (ONC), and online regression (ONR)-are considered, categorized by the existence of online update and the types of supervision signal. To account for the problem, we resort to the idea of ensemble and propose a dual-modal framework for target localization, consisting of robust localization suppressing distractors via ONR and the accurate localization attending to the target center precisely via OFC. To yield a final representation (i.e, bounding box), we propose a simple but effective score voting strategy to involve adjacent predictions such that the final representation does not commit to a single location. Operating beyond the real-time demand, our proposed method is further validated on 8 datasets-VOT2018, VOT2019, OTB2015, NFS, UAV123, LaSOT, TrackingNet, and GOT-10k, achieving state-of-the-art performance.

cs.CV↗

arXiv2021

Learning Collision-Free Space Detection from Stereo Images: Homography Matrix Brings Better Data Augmentation

Collision-free space detection is a critical component of autonomous vehicle perception. The state-of-the-art algorithms are typically based on supervised learning. The performance of such approaches is always dependent on the quality and amount of labeled training data. Additionally, it remains an open challenge to train deep convolutional neural networks (DCNNs) using only a small quantity of training samples. Therefore, this paper mainly explores an effective training data augmentation approach that can be employed to improve the overall DCNN performance, when additional images captured from different views are available. Due to the fact that the pixels of the collision-free space (generally regarded as a planar surface) between two images captured from different views can be associated by a homography matrix, the scenario of the target image can be transformed into the reference view. This provides a simple but effective way of generating training data from additional multi-view images. Extensive experimental results, conducted with six state-of-the-art semantic segmentation DCNNs on three datasets, demonstrate the effectiveness of our proposed training data augmentation algorithm for enhancing collision-free space detection performance. When validated on the KITTI road benchmark, our approach provides the best results for stereo vision-based collision-free space detection.

cs.RO↗

arXiv2021

ORBBuf: A Robust Buffering Method for Remote Visual SLAM

The data loss caused by unreliable network seriously impacts the results of remote visual SLAM systems. From our experiment, a loss of less than 1 second of data can cause a visual SLAM algorithm to lose tracking. We present a novel buffering method, ORBBuf, to reduce the impact of data loss on remote visual SLAM systems. We model the buffering problem as an optimization problem by introducing a similarity metric between frames. To solve the buffering problem, we present an efficient greedy-like algorithm to discard the frames that have the least impact on the quality of SLAM results. We implement our ORBBuf method on ROS, a widely used middleware framework. Through an extensive evaluation on real-world scenarios and tens of gigabytes of datasets, we demonstrate that our ORBBuf method can be applied to different state-estimation algorithms (DSO and VINS-Fusion), different sensor data (both monocular images and stereo images), different scenes (both indoor and outdoor), and different network environments (both WiFi networks and 4G networks). Our experimental results indicate that the network losses indeed affect the SLAM results, and our ORBBuf method can reduce the RMSE up to 50 times comparing with the Drop-Oldest and Random buffering methods.

cs.RO↗

arXiv2021

STUaNet: Understanding uncertainty in spatiotemporal collective human mobility

The high dynamics and heterogeneous interactions in the complicated urban systems have raised the issue of uncertainty quantification in spatiotemporal human mobility, to support critical decision-makings in risk-aware web applications such as urban event prediction where fluctuations are of significant interests. Given the fact that uncertainty quantifies the potential variations around prediction results, traditional learning schemes always lack uncertainty labels, and conventional uncertainty quantification approaches mostly rely upon statistical estimations with Bayesian Neural Networks or ensemble methods. However, they have never involved any spatiotemporal evolution of uncertainties under various contexts, and also have kept suffering from the poor efficiency of statistical uncertainty estimation while training models with multiple times. To provide high-quality uncertainty quantification for spatiotemporal forecasting, we propose an uncertainty learning mechanism to simultaneously estimate internal data quality and quantify external uncertainty regarding various contextual interactions. To address the issue of lacking labels of uncertainty, we propose a hierarchical data turbulence scheme where we can actively inject controllable uncertainty for guidance, and hence provide insights to both uncertainty quantification and weak supervised learning. Finally, we re-calibrate and boost the prediction performance by devising a gated-based bridge to adaptively leverage the learned uncertainty into predictions. Extensive experiments on three real-world spatiotemporal mobility sets have corroborated the superiority of our proposed model in terms of both forecasting and uncertainty quantification.

cs.LG↗

arXiv2021

Optimal Sequential Tests for Detection of Changes under Finite measure space for Finite Sequences of Networks

This paper considers the change-point problem for finite sequences of networks. To avoid the difficulty of computing the normalization coefficient, such as in Exponential random graphical models (ERGMs) and Markov networks, we construct a finite measure space with measure ratio statistics. A new performance measure of detection delay is proposed to detect the changes in distribution of the network. And an optimal sequential test is proposed under the performance measure. The good performance of the optimal sequential test is illustrated numerically on ERGMs and Erdos-Rényi network sequences.

math.ST↗

arXiv2020

Enhanced anisotropic superconductivity in the topological nodal-line semimetal InxTaS2

Coexistence of topological bands and charge density wave (CDW) in topological materials has attracted immense attentions because of their fantastic properties, such as axionic-CDW, three-dimensional quantum Hall effect, etc. In this work, a nodal-line semimetal InxTaS2 characterized by CDW and superconductivity is successfully synthesized, whose structure and topological bands (two separated Wely rings) are similar to In0.58TaSe2. A 2 x 2 commensurate CDW is observed at low temperature in InxTaS2, identified by transport properties and STM measurements. Moreover, superconductivity emerges below 0.69 K, and the anisotropy ratio of upper critical field [Gamma = H||ab c2(0)=H||c c2(0)] is significantly enhanced compared to 2H-TaS2, which shares the same essential layer unit. According to the Lawrence-Doniach model, the enhanced Gamma may be explained by the reduced effective mass in kx-ky plane, where Weyl rings locate. Therefore, this type of layered topological systems may offer a platform to investigate highly anisotropic superconductivity and to understand the extremely large upper critical field in the bulk or in the two-dimensional limit.

cond-mat.supr-con↗

arXiv2020

Type-I superconductivity in noncentrosymmetric NbGe$_{2}$

Single crystals of NbGe$_{2}$ which crystallize in a noncentrosymmetric hexagonal structure with chirality are synthesized and their superconductivity is investigated. Type-I superconductivity is confirmed by dc magnetization, field-induced second-to first-order phase transition in specific heat, and a small Ginzburg-Landau parameter $κ_{GL}=0.12$. The isothermal magnetization measurements show that there is a crossover from type-I to type-II/1 superconductivity with decreasing temperature and an unusually enhanced surface superconducting critical field ($H_{c3}$) is discovered. The band structure calculations indicate the presence of Kramer-Weyl nodes near the Fermi level. These observations demonstrate that NbGe$_{2}$ is an interesting and rare example involving the possible interplay of type-I superconductivity, noncentrosymmetric structure and topological properties.

cond-mat.supr-con↗

arXiv2020

Astraea: Self-balancing Federated Learning for Improving Classification Accuracy of Mobile Deep Learning Applications

Federated learning (FL) is a distributed deep learning method which enables multiple participants, such as mobile phones and IoT devices, to contribute a neural network model while their private training data remains in local devices. This distributed approach is promising in the edge computing system where have a large corpus of decentralized data and require high privacy. However, unlike the common training dataset, the data distribution of the edge computing system is imbalanced which will introduce biases in the model training and cause a decrease in accuracy of federated learning applications. In this paper, we demonstrate that the imbalanced distributed training data will cause accuracy degradation in FL. To counter this problem, we build a self-balancing federated learning framework call Astraea, which alleviates the imbalances by 1) Global data distribution based data augmentation, and 2) Mediator based multi-client rescheduling. The proposed framework relieves global imbalance by runtime data augmentation, and for averaging the local imbalance, it creates the mediator to reschedule the training of clients based on Kullback-Leibler divergence (KLD) of their data distribution. Compared with FedAvg, the state-of-the-art FL algorithm, Astraea shows +5.59% and +5.89% improvement of top-1 accuracy on the imbalanced EMNIST and imbalanced CINIC-10 datasets, respectively. Meanwhile, the communication traffic of Astraea can be 82% lower than that of FedAvg.

cs.LG↗

arXiv2020

Anisotropic gapping of topological Weyl rings in the charge-density-wave superconductor InxTaSe2

Topological materials and topological phases have recently become a hot topic in condensed matter physics. In this work, we report a topological nodal-line semimetal InxTaSe2, in the presence of both charge density wave (CDW) and superconductivity. In the x = 0.58 samples, the 2 * /3 commensurate CDW (CCDW) and the 2 * 2 CCDW are observed below 116 K and 77 K, respectively. Consistent with theoretical calculations, the spin-orbital coupling gives rise to two two-fold-degenerate nodal rings (Weyl rings) connected by drumhead surface states, confirmed by angle-resolved photoemission spectroscopy. Our results suggest that the 2 * 2 CCDW ordering gaps out one Weyl ring in accordance with the CDW band folding, while the other Weyl ring remains gapless with intact surface states. In addition, superconductivity emerges at 0.91 K, with the upper critical field deviating from the s-wave behavior at low temperature, implying possibly unconventional superconductivity. Therefore, InxTaSe2 represents an interesting material system to study the interplay between CDW, nontrivial band topology and superconductivity.

cond-mat.supr-con↗

arXiv2020

A new version of a theorem of Kaplansky

A well-known theorem of Kaplansky states that any projective module is a direct sum of countably generated modules. In this paper, we prove the $w$-version of this theorem, where $w$ is a hereditary torsion theory for modules over a commutative ring.

math.AC↗

arXiv2019

Universal trapping law induced by atomic cloud in single-photon cooperative dynamics

Single-photon cooperative dynamics of an assembly of two-level quantum emitters coupled by a bosonic bath are investigated. The bosonic bath is general and it can be anything as long as the exchange of excitations between quantum emitters and bath is present. In these systems, it is found that the population on the excited emitter keeps a simple and universal trapping law due to the existence of system's dark states. Different from the trapping regime caused by photonemitter dressed states, this type of trapping is only associated with the number of quantum emitters. According to the trapping law, the cooperative spontaneous emission at single-photon level in this kind of systems is universally inhibited when the emitter number is large enough.

quant-ph↗

arXiv2019

PT-ResNet: Perspective Transformation-Based Residual Network for Semantic Road Image Segmentation

Semantic road region segmentation is a high-level task, which paves the way towards road scene understanding. This paper presents a residual network trained for semantic road segmentation. Firstly, we represent the projections of road disparities in the v-disparity map as a linear model, which can be estimated by optimizing the v-disparity map using dynamic programming. This linear model is then utilized to reduce the redundant information in the left and right road images. The right image is also transformed into the left perspective view, which greatly enhances the road surface similarity between the two images. Finally, the processed stereo images and their disparity maps are concatenated to create a set of 3D images, which are then utilized to train our neural network. The experimental results illustrate that our network achieves a maximum F1-measure of approximately 91.19% when analyzing the images from the KITTI road dataset.

cs.CV↗

arXiv2019

Autonomous UAV Landing System Based on Visual Navigation

In this paper, we present an autonomous unmanned aerial vehicle (UAV) landing system based on visual navigation. We design the landmark as a topological pattern in order to enable the UAV to distinguish the landmark from the environment easily. In addition, a dynamic thresholding method is developed for image binarization to improve detection efficiency. The relative distance in the horizontal plane is calculated according to effective image information, and the relative height is obtained using a linear interpolation method. The landing experiments are performed on a static and a moving platform, respectively. The experimental results illustrate that our proposed landing system performs robustly and accurately.

cs.RO↗