arXiv subjects

Wei Feng

Publications and source records attributed to Wei Feng.

At least 217 records · Page 12Linked to original sources

arXiv2022

Regularized Modal Regression on Markov-dependent Observations: A Theoretical Assessment

Modal regression, a widely used regression protocol, has been extensively investigated in statistical and machine learning communities due to its robustness to outliers and heavy-tailed noises. Understanding modal regression's theoretical behavior can be fundamental in learning theory. Despite significant progress in characterizing its statistical property, the majority of the results are based on the assumption that samples are independent and identical distributed (i.i.d.), which is too restrictive for real-world applications. This paper concerns the statistical property of regularized modal regression (RMR) within an important dependence structure - Markov dependent. Specifically, we establish the upper bound for RMR estimator under moderate conditions and give an explicit learning rate. Our results show that the Markov dependence impacts on the generalization error in the way that sample size would be discounted by a multiplicative factor depending on the spectral gap of underlying Markov chain. This result shed a new light on characterizing the theoretical underpinning for robust regression.

stat.ML↗

arXiv2022

MISF: Multi-level Interactive Siamese Filtering for High-Fidelity Image Inpainting

Although achieving significant progress, existing deep generative inpainting methods are far from real-world applications due to the low generalization across different scenes. As a result, the generated images usually contain artifacts or the filled pixels differ greatly from the ground truth. Image-level predictive filtering is a widely used image restoration technique, predicting suitable kernels adaptively according to different input scenes. Inspired by this inherent advantage, we explore the possibility of addressing image inpainting as a filtering task. To this end, we first study the advantages and challenges of image-level predictive filtering for image inpainting: the method can preserve local structures and avoid artifacts but fails to fill large missing areas. Then, we propose semantic filtering by conducting filtering on the deep feature level, which fills the missing semantic information but fails to recover the details. To address the issues while adopting the respective advantages, we propose a novel filtering technique, i.e., Multilevel Interactive Siamese Filtering (MISF), which contains two branches: kernel prediction branch (KPB) and semantic & image filtering branch (SIFB). These two branches are interactively linked: SIFB provides multi-level features for KPB while KPB predicts dynamic kernels for SIFB. As a result, the final method takes the advantage of effective semantic & image-level filling for high-fidelity inpainting. We validate our method on three challenging datasets, i.e., Dunhuang, Places2, and CelebA. Our method outperforms state-of-the-art baselines on four metrics, i.e., L1, PSNR, SSIM, and LPIPS. Please try the released code and model at https://github.com/tsingqguo/misf.

cs.CV↗

arXiv2022

AGCN: Augmented Graph Convolutional Network for Lifelong Multi-label Image Recognition

The Lifelong Multi-Label (LML) image recognition builds an online class-incremental classifier in a sequential multi-label image recognition data stream. The key challenges of LML image recognition are the construction of label relationships on Partial Labels of training data and the Catastrophic Forgetting on old classes, resulting in poor generalization. To solve the problems, the study proposes an Augmented Graph Convolutional Network (AGCN) model that can construct the label relationships across the sequential recognition tasks and sustain the catastrophic forgetting. First, we build an Augmented Correlation Matrix (ACM) across all seen classes, where the intra-task relationships derive from the hard label statistics while the inter-task relationships leverage both hard and soft labels from data and a constructed expert network. Then, based on the ACM, the proposed AGCN captures label dependencies with dynamic augmented structure and yields effective class representations. Last, to suppress the forgetting of label dependencies across old tasks, we propose a relationship-preserving loss as a constraint to the construction of label relationships. The proposed method is evaluated using two multi-label image benchmarks and the experimental results show that the proposed method is effective for LML image recognition and can build convincing correlation across tasks even if the labels of previous tasks are missing. Our code is available at https://github.com/Kaile-Du/AGCN.

cs.CV↗

arXiv2022

Adversarial Rain Attack and Defensive Deraining for DNN Perception

Rain often poses inevitable threats to deep neural network (DNN) based perception systems, and a comprehensive investigation of the potential risks of the rain to DNNs is of great importance. However, it is rather difficult to collect or synthesize rainy images that can represent all rain situations that would possibly occur in the real world. To this end, in this paper, we start from a new perspective and propose to combine two totally different studies, i.e., rainy image synthesis and adversarial attack. We first present an adversarial rain attack, with which we could simulate various rain situations with the guidance of deployed DNNs and reveal the potential threat factors that can be brought by rain. In particular, we design a factor-aware rain generation that synthesizes rain streaks according to the camera exposure process and models the learnable rain factors for adversarial attack. With this generator, we perform the adversarial rain attack against the image classification and object detection. To defend the DNNs from the negative rain effect, we also present a defensive deraining strategy, for which we design an adversarial rain augmentation that uses mixed adversarial rain layers to enhance deraining models for downstream DNN perception. Our large-scale evaluation on various datasets demonstrates that our synthesized rainy images with realistic appearances not only exhibit strong adversarial capability against DNNs, but also boost the deraining models for defensive purposes, building the foundation for further rain-robust perception studies.

cs.CV↗

arXiv2022

Uncertainty-Aware Cascaded Dilation Filtering for High-Efficiency Deraining

Deraining is a significant and fundamental computer vision task, aiming to remove the rain streaks and accumulations in an image or video captured under a rainy day. Existing deraining methods usually make heuristic assumptions of the rain model, which compels them to employ complex optimization or iterative refinement for high recovery quality. This, however, leads to time-consuming methods and affects the effectiveness for addressing rain patterns deviated from from the assumptions. In this paper, we propose a simple yet efficient deraining method by formulating deraining as a predictive filtering problem without complex rain model assumptions. Specifically, we identify spatially-variant predictive filtering (SPFilt) that adaptively predicts proper kernels via a deep network to filter different individual pixels. Since the filtering can be implemented via well-accelerated convolution, our method can be significantly efficient. We further propose the EfDeRain+ that contains three main contributions to address residual rain traces, multi-scale, and diverse rain patterns without harming the efficiency. First, we propose the uncertainty-aware cascaded predictive filtering (UC-PFilt) that can identify the difficulties of reconstructing clean pixels via predicted kernels and remove the residual rain traces effectively. Second, we design the weight-sharing multi-scale dilated filtering (WS-MS-DFilt) to handle multi-scale rain streaks without harming the efficiency. Third, to eliminate the gap across diverse rain patterns, we propose a novel data augmentation method (i.e., RainMix) to train our deep models. By combining all contributions with sophisticated analysis on different variants, our final method outperforms baseline methods on four single-image deraining datasets and one video deraining dataset in terms of both recovery quality and speed.

cs.CV↗

arXiv2021

Benchmarking Shadow Removal for Facial Landmark Detection and Beyond

Facial landmark detection is a very fundamental and significant vision task with many important applications. In practice, facial landmark detection can be affected by a lot of natural degradations. One of the most common and important degradations is the shadow caused by light source blocking. While many advanced shadow removal methods have been proposed to recover the image quality in recent years, their effects to facial landmark detection are not well studied. For example, it remains unclear whether shadow removal could enhance the robustness of facial landmark detection to diverse shadow patterns or not. In this work, for the first attempt, we construct a novel benchmark to link two independent but related tasks (i.e., shadow removal and facial landmark detection). In particular, the proposed benchmark covers diverse face shadows with different intensities, sizes, shapes, and locations. Moreover, to mine hard shadow patterns against facial landmark detection, we propose a novel method (i.e., adversarial shadow attack), which allows us to construct a challenging subset of the benchmark for a comprehensive analysis. With the constructed benchmark, we conduct extensive analysis on three state-of-the-art shadow removal methods and three landmark detectors. The observation of this work motivates us to design a novel detection-aware shadow removal framework, which empowers shadow removal to achieve higher restoration quality and enhance the shadow robustness of deployed facial landmark detectors.

cs.CV↗

arXiv2021

Cross-correlation mediated by Majorana island with finite charging energy

Based on the many-particle-number-state treatment for transport through a pair of Majorana zero modes (MZMs) which are coupled to the leads via two quantum dots, we identify that the reason for zero cross correlation of currents at uncoupling limit between the MZMs is from a degeneracy of the teleportation and the Andreev process channels. We then propose a scheme to eliminate the degeneracy by introducing finite charging energy on the Majorana island which allows for coexistence of the two channels. We find nonzero cross correlation established even in the Majorana uncoupling limit (and also in the small charging energy limit), which demonstrates well the teleportation or nonlocal nature of the MZMs. More specifically, the characteristic structure of coherent peaks in the power spectrum of the cross correlation is analyzed to identify the nonlocal and coherent coupling mechanism between the MZMs and the quantum dots. We also display the behaviors of peak shift with variation of the Majorana coupling energy, which can be realized by modulating parameters such as the magnetic field.

cond-mat.mes-hall↗

arXiv2021

Local and nonlocal complex discrete and semi-discrete sine-Gordon equations and solutions

In this paper, local and nonlocal complex reduction of a discrete and a semi-discrete negative order Ablowitz-Kaup-Newell- Segur equations is studied. Cauchy matrix type solutions, including soliton solutions and Jordan-block solutions, for the resulting local and nonlocal complex discrete and semi-discrete sine-Gordon equations are constructed. Dynamics of 1-soliton solution are analyzed and illustrated.

nlin.SI↗

arXiv2021

Solutions to nonlocal nonisospectral (2+1)-dimensional breaking soliton equations

Nonlocal reductions of a nonisospectral (2+1)-dimensional breaking soliton Ablowitz-Kaup-Newell-Segur equation are discussed on the base of double Wronskian reduction technique. Various types of solutions, including soliton solutions and Jordan-block solutions, for the resulting nonlocal equations are derived. Dynamics of these obtained solutions are analyzed and illustrated.

nlin.SI↗

arXiv2021

Majorana Conductances in Three-Terminal Transports

We consider a two-lead (three-terminal) setup of nonlocal transport through Majorana zero modes (MZMs) and construct a Majorana master equation (which is also valid for small bias voltage). We first carry out representative results of current and then show that a modified Bogoliubov-de Gennes (BdG) treatment can consistently recover the same results. Based on the interplay of the two approaches, we reveal the existence of nonvanishing channels of teleportation and crossed Andreev reflections even at the limit $ε_M\to 0$ (zero coupling energy of the MZMs), which leads to new predictions for the height of the zero-bias-peak of the local conductance and the $ε_M$-scaling behavior of the teleportation conductance, for verification by experiments.

cond-mat.mes-hall↗

arXiv2021

Pasadena: Perceptually Aware and Stealthy Adversarial Denoise Attack

Image denoising can remove natural noise that widely exists in images captured by multimedia devices due to low-quality imaging sensors, unstable image transmission processes, or low light conditions. Recent works also find that image denoising benefits the high-level vision tasks, e.g., image classification. In this work, we try to challenge this common sense and explore a totally new problem, i.e., whether the image denoising can be given the capability of fooling the state-of-the-art deep neural networks (DNNs) while enhancing the image quality. To this end, we initiate the very first attempt to study this problem from the perspective of adversarial attack and propose the adversarial denoise attack. More specifically, our main contributions are three-fold: First, we identify a new task that stealthily embeds attacks inside the image denoising module widely deployed in multimedia devices as an image post-processing operation to simultaneously enhance the visual image quality and fool DNNs. Second, we formulate this new task as a kernel prediction problem for image filtering and propose the adversarial-denoising kernel prediction that can produce adversarial-noiseless kernels for effective denoising and adversarial attacking simultaneously. Third, we implement an adaptive perceptual region localization to identify semantic-related vulnerability regions with which the attack can be more effective while not doing too much harm to the denoising. We name the proposed method as Pasadena (Perceptually Aware and Stealthy Adversarial DENoise Attack) and validate our method on the NeurIPS'17 adversarial competition dataset, CVPR2021-AIC-VI: unrestricted adversarial attacks on ImageNet,etc. The comprehensive evaluation and analysis demonstrate that our method not only realizes denoising but also achieves a significantly higher success rate and transferability over state-of-the-art attacks.

cs.CV↗

arXiv2021

VIL-100: A New Dataset and A Baseline Model for Video Instance Lane Detection

Lane detection plays a key role in autonomous driving. While car cameras always take streaming videos on the way, current lane detection works mainly focus on individual images (frames) by ignoring dynamics along the video. In this work, we collect a new video instance lane detection (VIL-100) dataset, which contains 100 videos with in total 10,000 frames, acquired from different real traffic scenarios. All the frames in each video are manually annotated to a high-quality instance-level lane annotation, and a set of frame-level and video-level metrics are included for quantitative performance evaluation. Moreover, we propose a new baseline model, named multi-level memory aggregation network (MMA-Net), for video instance lane detection. In our approach, the representation of current frame is enhanced by attentively aggregating both local and global memory features from other frames. Experiments on the new collected dataset show that the proposed MMA-Net outperforms state-of-the-art lane detection methods and video object segmentation methods. We release our dataset and code at https://github.com/yujun0-0/MMA-Net.

cs.CV↗

arXiv2021

NOMA-Based Hybrid Satellite-UAV-Terrestrial Networks for Beyond 5G Maritime Internet of Things

Current fifth-generation (5G) networks do not cover maritime areas, causing difficulties in developing maritime Internet of Things (IoT). To tackle this problem, we establish a nearshore network by collaboratively using on-shore terrestrial base stations (TBSs) and tethered unmanned aerial vehicles (UAVs). These TBSs and UAVs form virtual clusters in a user-centric manner. Within each virtual cluster, non-orthogonal multiple access (NOMA) is adopted for agilely including various maritime IoT devices, which are usually sparsely distributed on the vast ocean. The nearshore network also shares spectrum with marine satellites. In such a NOMA-based hybrid satellite-UAV-terrestrial network, interference among different network segments, different clusters, as well as different users occurs. We thereby formulate a joint power allocation problem to maximize the sum rate of the network. Different from existing studies, we use large-scale channel state information (CSI) only for optimization to reduce system overhead. The large-scale CSI is obtained by using the position information of maritime IoT devices. The problem is non-convex with intractable non-linear constraints. We tackle these difficulties by adopting the max-min optimization, auxiliary function method, and successive convex approximation technique. An iterative power allocation algorithm is accordingly proposed, which is shown effective for coverage enhancement by simulations. This shows the potential of NOMA-based hybrid satellite-UAV-terrestrial networks for maritime on-demand coverage.

cs.IT↗

arXiv2021

From Synthetic to Real: Image Dehazing Collaborating with Unlabeled Real Data

Single image dehazing is a challenging task, for which the domain shift between synthetic training data and real-world testing images usually leads to degradation of existing methods. To address this issue, we propose a novel image dehazing framework collaborating with unlabeled real data. First, we develop a disentangled image dehazing network (DID-Net), which disentangles the feature representations into three component maps, i.e. the latent haze-free image, the transmission map, and the global atmospheric light estimate, respecting the physical model of a haze process. Our DID-Net predicts the three component maps by progressively integrating features across scales, and refines each map by passing an independent refinement network. Then a disentangled-consistency mean-teacher network (DMT-Net) is employed to collaborate unlabeled real data for boosting single image dehazing. Specifically, we encourage the coarse predictions and refinements of each disentangled component to be consistent between the student and teacher networks by using a consistency loss on unlabeled real data. We make comparison with 13 state-of-the-art dehazing methods on a new collected dataset (Haze4K) and two widely-used dehazing datasets (i.e., SOTS and HazeRD), as well as on real-world hazy images. Experimental results demonstrate that our method has obvious quantitative and qualitative improvements over the existing methods.

cs.CV↗

arXiv2021

Unsupervised Domain Adaptation for Retinal Vessel Segmentation with Adversarial Learning and Transfer Normalization

Retinal vessel segmentation plays a key role in computer-aided screening, diagnosis, and treatment of various cardiovascular and ophthalmic diseases. Recently, deep learning-based retinal vessel segmentation algorithms have achieved remarkable performance. However, due to the domain shift problem, the performance of these algorithms often degrades when they are applied to new data that is different from the training data. Manually labeling new data for each test domain is often a time-consuming and laborious task. In this work, we explore unsupervised domain adaptation in retinal vessel segmentation by using entropy-based adversarial learning and transfer normalization layer to train a segmentation network, which generalizes well across domains and requires no annotation of the target domain. Specifically, first, an entropy-based adversarial learning strategy is developed to reduce the distribution discrepancy between the source and target domains while also achieving the objective of entropy minimization on the target domain. In addition, a new transfer normalization layer is proposed to further boost the transferability of the deep network. It normalizes the features of each domain separately to compensate for the domain distribution gap. Besides, it also adaptively selects those feature channels that are more transferable between domains, thus further enhancing the generalization performance of the network. We conducted extensive experiments on three regular fundus image datasets and an ultra-widefield fundus image dataset, and the results show that our approach yields significant performance gains compared to other state-of-the-art methods.

eess.IV↗

arXiv2021

CarveNet: Carving Point-Block for Complex 3D Shape Completion

3D point cloud completion is very challenging because it heavily relies on the accurate understanding of the complex 3D shapes (e.g., high-curvature, concave/convex, and hollowed-out 3D shapes) and the unknown & diverse patterns of the partially available point clouds. In this paper, we propose a novel solution,i.e., Point-block Carving (PC), for completing the complex 3D point cloud completion. Given the partial point cloud as the guidance, we carve a3D block that contains the uniformly distributed 3D points, yielding the entire point cloud. To achieve PC, we propose a new network architecture, i.e., CarveNet. This network conducts the exclusive convolution on each point of the block, where the convolutional kernels are trained on the 3D shape data. CarveNet determines which point should be carved, for effectively recovering the details of the complete shapes. Furthermore, we propose a sensor-aware method for data augmentation,i.e., SensorAug, for training CarveNet on richer patterns of partial point clouds, thus enhancing the completion power of the network. The extensive evaluations on the ShapeNet and KITTI datasets demonstrate the generality of our approach on the partial point clouds with diverse patterns. On these datasets, CarveNet successfully outperforms the state-of-the-art methods.

cs.CV↗

arXiv2021

MEC-Empowered Non-Terrestrial Network for 6G Wide-Area Time-Sensitive Internet of Things

In the upcoming sixth-generation (6G) era, the demand for constructing a wide-area time-sensitive Internet of Things (IoT) keeps increasing. As conventional cellular technologies are hard to be directly used for wide-area time-sensitive IoT, it is beneficial to use non-terrestrial infrastructures including satellites and unmanned aerial vehicles (UAVs), where a non-terrestrial network (NTN) can be built under the cell-free architecture. Driven by the time-sensitive requirements and uneven distribution of IoT devices, the NTN is required to be empowered by mobile edge computing (MEC) while providing oasis-oriented on-demand coverage for the devices. Nevertheless, communication and MEC systems are coupled with each other under the influence of complex propagation environment in the MEC-empowered NTN, which makes it hard to orchestrate the resources. In this paper, we propose a process-oriented framework to design the communication and MEC systems in a time-division manner. Under this framework, the large-scale channel state information (CSI) is used to characterize the complex propagation environment with an affordable cost, where a non-convex latency minimization problem is formulated. After that, the approximated problem is given and it can be decomposed into subproblems. These subproblems are further solved in an iterative way. Simulation results demonstrate the superiority of the proposed process-oriented scheme over other algorithms. These results also indicate that the payload deployments of UAVs should be appropriately predesigned to improve the efficiency of resource use. Furthermore, the results imply that it is advantageous to integrate NTN with MEC for wide-area time-sensitive IoT.

eess.SP↗

arXiv2021

Physical Layer Security for UAV Communications in 5G and Beyond Networks

Due to its high mobility and flexible deployment, unmanned aerial vehicle (UAV) is drawing unprecedented interest in both military and civil applications to enable agile wireless communications and provide ubiquitous connectivity. Mainly operating in an open environment, UAV communications can benefit from dominant line-of-sight links; however, it on the other hand renders the UAVs more vulnerable to malicious eavesdropping or jamming attacks. Recently, physical layer security (PLS), which exploits the inherent randomness of the wireless channels for secure communications, has been introduced to UAV systems as an important complement to the conventional cryptography-based approaches. In this paper, a comprehensive survey on the current achievements of the UAV-aided wireless communications is conducted from the PLS perspective. We first introduce the basic concepts of UAV communications including the typical static/mobile deployment scenarios, the unique characteristics of air-to-ground channels, as well as various roles that a UAV may act when PLS is concerned. Then, we introduce the widely used secrecy performance metrics and start by reviewing the secrecy performance analysis and enhancing techniques for statically deployed UAV systems, and extend the discussion to a more general scenario where the UAVs' mobility is further exploited. For both cases, respectively, we summarize the commonly adopted methodologies in the corresponding analysis and design, then describe important works in the literature in detail. Finally, potential research directions and challenges are discussed to provide an outlook for future works in the area of UAV-PLS in 5G and beyond networks.

cs.NI↗