arXiv subjects

Lu Zhang

Publications and source records attributed to Lu Zhang.

At least 361 records · Page 20Linked to original sources

arXiv2021

Transformer-based Network for RGB-D Saliency Detection

RGB-D saliency detection integrates information from both RGB images and depth maps to improve prediction of salient regions under challenging conditions. The key to RGB-D saliency detection is to fully mine and fuse information at multiple scales across the two modalities. Previous approaches tend to apply the multi-scale and multi-modal fusion separately via local operations, which fails to capture long-range dependencies. Here we propose a transformer-based network to address this issue. Our proposed architecture is composed of two modules: a transformer-based within-modality feature enhancement module (TWFEM) and a transformer-based feature fusion module (TFFM). TFFM conducts a sufficient feature fusion by integrating features from multiple scales and two modalities over all positions simultaneously. TWFEM enhances feature on each scale by selecting and integrating complementary information from other scales within the same modality before TFFM. We show that transformer is a uniform operation which presents great efficacy in both feature fusion and feature enhancement, and simplifies the model design. Extensive experimental results on six benchmark datasets demonstrate that our proposed network performs favorably against state-of-the-art RGB-D saliency detection methods.

cs.CV↗

arXiv2021

SAFA: Structure Aware Face Animation

Recent success of generative adversarial networks (GAN) has made great progress on the face animation task. However, the complex scene structure of a face image still makes it a challenge to generate videos with face poses significantly deviating from the source image. On one hand, without knowing the facial geometric structure, generated face images might be improperly distorted. On the other hand, some area of the generated image might be occluded in the source image, which makes it difficult for GAN to generate realistic appearance. To address these problems, we propose a structure aware face animation (SAFA) method which constructs specific geometric structures to model different components of a face image. Following the well recognized motion based face animation technique, we use a 3D morphable model (3DMM) to model the face, multiple affine transforms to model the other foreground components like hair and beard, and an identity transform to model the background. The 3DMM geometric embedding not only helps generate realistic structure for the driving scene, but also contributes to better perception of occluded area in the generated image. Besides, we further propose to exploit the widely studied inpainting technique to faithfully recover the occluded image area. Both quantitative and qualitative experiment results have shown the superiority of our method. Code is available at https://github.com/Qiulin-W/SAFA.

cs.CV↗

arXiv2021

New insights in smoothness and strong convexity with improved convergence of gradient descent

The starting assumptions to study the convergence and complexity of gradient-type methods may be the smoothness (also called Lipschitz continuity of gradient) and the strong convexity. In this note, we revisit these two basic properties from a new perspective that motivates their definitions and equivalent characterizations, along with an improved linear convergence of the gradient descent method.

math.OC↗

arXiv2021

DP-SSL: Towards Robust Semi-supervised Learning with A Few Labeled Samples

The scarcity of labeled data is a critical obstacle to deep learning. Semi-supervised learning (SSL) provides a promising way to leverage unlabeled data by pseudo labels. However, when the size of labeled data is very small (say a few labeled samples per class), SSL performs poorly and unstably, possibly due to the low quality of learned pseudo labels. In this paper, we propose a new SSL method called DP-SSL that adopts an innovative data programming (DP) scheme to generate probabilistic labels for unlabeled data. Different from existing DP methods that rely on human experts to provide initial labeling functions (LFs), we develop a multiple-choice learning~(MCL) based approach to automatically generate LFs from scratch in SSL style. With the noisy labels produced by the LFs, we design a label model to resolve the conflict and overlap among the noisy labels, and finally infer probabilistic labels for unlabeled samples. Extensive experiments on four standard SSL benchmarks show that DP-SSL can provide reliable labels for unlabeled data and achieve better classification performance on test sets than existing SSL methods, especially when only a small number of labeled samples are available. Concretely, for CIFAR-10 with only 40 labeled samples, DP-SSL achieves 93.82% annotation accuracy on unlabeled data and 93.46% classification accuracy on test data, which are higher than the SOTA results.

cs.CV↗

arXiv2021

Learning Synergistic Attention for Light Field Salient Object Detection

We propose a novel Synergistic Attention Network (SA-Net) to address the light field salient object detection by establishing a synergistic effect between multi-modal features with advanced attention mechanisms. Our SA-Net exploits the rich information of focal stacks via 3D convolutional neural networks, decodes the high-level features of multi-modal light field data with two cascaded synergistic attention modules, and predicts the saliency map using an effective feature fusion module in a progressive manner. Extensive experiments on three widely-used benchmark datasets show that our SA-Net outperforms 28 state-of-the-art models, sufficiently demonstrating its effectiveness and superiority. Our code is available at https://github.com/PanoAsh/SA-Net.

cs.CV↗

arXiv2021

Charge instability of topological Fermi arcs in chiral crystal CoSi

Topological boundary states, emerged at the spatial boundary between topological non-trivial and trivial phases, are usually gapless, or commonly referred as metallic states. For example, the surface state of a topological insulator is a gapless Dirac state. These metallic topological boundary states are typically well described by non-interacting fermions. However, the behavior of topological boundary states with significant electron-electron interactions, which could turn the gapless boundary states into gapped ordered states, e.g., density wave states or superconducting states, is of great interest theoretically, but is still lacking evidence experimentally. Here, we report the observation of incommensurable charge density wave (CDW) formed on the topological boundary states driven by the electron-electron interactions on the (001) surface of CoSi. The wavevector of CDW varies as the temperature changes, which coincides with the evolution of topological surface Fermi arcs with temperature. The orientation of CDW phase is determined by the chirality of the Fermi arcs, which indicates direct association between CDW and Fermi arcs. Our finding will stimulate the search of more interactions-driven ordered states, such as superconductivity and magnetism, on the boundaries of topological materials.

cond-mat.str-el↗

arXiv2021

Weakly-supervised Text Classification Based on Keyword Graph

Weakly-supervised text classification has received much attention in recent years for it can alleviate the heavy burden of annotating massive data. Among them, keyword-driven methods are the mainstream where user-provided keywords are exploited to generate pseudo-labels for unlabeled texts. However, existing methods treat keywords independently, thus ignore the correlation among them, which should be useful if properly exploited. In this paper, we propose a novel framework called ClassKG to explore keyword-keyword correlation on keyword graph by GNN. Our framework is an iterative process. In each iteration, we first construct a keyword graph, so the task of assigning pseudo labels is transformed to annotating keyword subgraphs. To improve the annotation quality, we introduce a self-supervised task to pretrain a subgraph annotator, and then finetune it. With the pseudo labels generated by the subgraph annotator, we then train a text classifier to classify the unlabeled texts. Finally, we re-extract keywords from the classified texts. Extensive experiments on both long-text and short-text datasets show that our method substantially outperforms the existing ones

cs.CL↗

arXiv2021

Achieving Counterfactual Fairness for Causal Bandit

In online recommendation, customers arrive in a sequential and stochastic manner from an underlying distribution and the online decision model recommends a chosen item for each arriving individual based on some strategy. We study how to recommend an item at each step to maximize the expected reward while achieving user-side fairness for customers, i.e., customers who share similar profiles will receive a similar reward regardless of their sensitive attributes and items being recommended. By incorporating causal inference into bandits and adopting soft intervention to model the arm selection strategy, we first propose the d-separation based UCB algorithm (D-UCB) to explore the utilization of the d-separation set in reducing the amount of exploration needed to achieve low cumulative regret. Based on that, we then propose the fair causal bandit (F-UCB) for achieving the counterfactual individual fairness. Both theoretical analysis and empirical evaluation demonstrate effectiveness of our algorithms.

cs.LG↗

arXiv2021

Hepatocellular Carcinoma Segmentation from Digital Subtraction Angiography Videos using Learnable Temporal Difference

Automatic segmentation of hepatocellular carcinoma (HCC) in Digital Subtraction Angiography (DSA) videos can assist radiologists in efficient diagnosis of HCC and accurate evaluation of tumors in clinical practice. Few studies have investigated HCC segmentation from DSA videos. It shows great challenging due to motion artifacts in filming, ambiguous boundaries of tumor regions and high similarity in imaging to other anatomical tissues. In this paper, we raise the problem of HCC segmentation in DSA videos, and build our own DSA dataset. We also propose a novel segmentation network called DSA-LTDNet, including a segmentation sub-network, a temporal difference learning (TDL) module and a liver region segmentation (LRS) sub-network for providing additional guidance. DSA-LTDNet is preferable for learning the latent motion information from DSA videos proactively and boosting segmentation performance. All of experiments are conducted on our self-collected dataset. Experimental results show that DSA-LTDNet increases the DICE score by nearly 4% compared to the U-Net baseline.

eess.IV↗

arXiv2021

A review of computational tools for generating metagenome-assembled genomes from metagenomic sequencing data

Microbes are essentially yet convolutedly linked with human lives on the earth. They critically interfere in different physiological processes and thus influence overall health status. Studying microbial species is used to be constrained to those that can be cultured in the lab. But it excluded a huge portion of the microbiome that could not survive on lab conditions. In the past few years, the culture-independent metagenomic sequencing enabled us to explore the complex microbial community coexisting within and on us. Metagenomics has equipped us with new avenues of investigating the microbiome, from studying a single species to a complex community in a dynamic ecosystem. Thus, identifying the involved microbes and their genomes becomes one of the core tasks in metagenomic sequencing. Metagenome-assembled genomes are groups of contigs with similar sequence characteristics from de novo assembly and could represent the microbial genomes from metagenomic sequencing. In this paper, we reviewed a spectrum of tools for producing and annotating metagenome-assembled genomes from metagenomic sequencing data and discussed their technical and biological perspectives.

q-bio.GN↗

arXiv2021

EPSILON: An Efficient Planning System for Automated Vehicles in Highly Interactive Environments

In this paper, we present an Efficient Planning System for automated vehicles In highLy interactive envirONments (EPSILON). EPSILON is an efficient interaction-aware planning system for automated driving, and is extensively validated in both simulation and real-world dense city traffic. It follows a hierarchical structure with an interactive behavior planning layer and an optimization-based motion planning layer. The behavior planning is formulated from a partially observable Markov decision process (POMDP), but is much more efficient than naively applying a POMDP to the decision-making problem. The key to efficiency is guided branching in both the action space and observation space, which decomposes the original problem into a limited number of closed-loop policy evaluations. Moreover, we introduce a new driver model with a safety mechanism to overcome the risk induced by the potential imperfectness of prior knowledge. For motion planning, we employ a spatio-temporal semantic corridor (SSC) to model the constraints posed by complex driving environments in a unified way. Based on the SSC, a safe and smooth trajectory is optimized, complying with the decision provided by the behavior planner. We validate our planning system in both simulations and real-world dense traffic, and the experimental results show that our EPSILON achieves human-like driving behaviors in highly interactive traffic flow smoothly and safely without being over-conservative compared to the existing planning methods.

cs.RO↗

arXiv2021

Medical-VLBERT: Medical Visual Language BERT for COVID-19 CT Report Generation With Alternate Learning

Medical imaging technologies, including computed tomography (CT) or chest X-Ray (CXR), are largely employed to facilitate the diagnosis of the COVID-19. Since manual report writing is usually too time-consuming, a more intelligent auxiliary medical system that could generate medical reports automatically and immediately is urgently needed. In this article, we propose to use the medical visual language BERT (Medical-VLBERT) model to identify the abnormality on the COVID-19 scans and generate the medical report automatically based on the detected lesion regions. To produce more accurate medical reports and minimize the visual-and-linguistic differences, this model adopts an alternate learning strategy with two procedures that are knowledge pretraining and transferring. To be more precise, the knowledge pretraining procedure is to memorize the knowledge from medical texts, while the transferring procedure is to utilize the acquired knowledge for professional medical sentences generations through observations of medical images. In practice, for automatic medical report generation on the COVID-19 cases, we constructed a dataset of 368 medical findings in Chinese and 1104 chest CT scans from The First Affiliated Hospital of Jinan University, Guangzhou, China, and The Fifth Affiliated Hospital of Sun Yat-sen University, Zhuhai, China. Besides, to alleviate the insufficiency of the COVID-19 training samples, our model was first trained on the large-scale Chinese CX-CHR dataset and then transferred to the COVID-19 CT dataset for further fine-tuning. The experimental results showed that Medical-VLBERT achieved state-of-the-art performances on terminology prediction and report generation with the Chinese COVID-19 CT dataset and the CX-CHR dataset. The Chinese COVID-19 CT dataset is available at https://covid19ct.github.io/.

cs.CV↗

arXiv2021

Semantic-driven Colorization

Recent colorization works implicitly predict the semantic information while learning to colorize black-and-white images. Consequently, the generated color is easier to be overflowed, and the semantic faults are invisible. As a human experience in colorization, our brains first detect and recognize the objects in the photo, then imagine their plausible colors based on many similar objects we have seen in real life, and finally colorize them, as described in the teaser. In this study, we simulate that human-like action to let our network first learn to understand the photo, then colorize it. Thus, our work can provide plausible colors at a semantic level. Plus, the semantic information of the learned model becomes understandable and able to interact. Additionally, we also prove that Instance Normalization is also a missing ingredient for colorization, then re-design the inference flow of U-Net to have two streams of data, providing an appropriate way of normalizing the feature maps from the black-and-white image and its semantic map. As a result, our network can provide plausible colors competitive to the typical colorization works for specific objects.

cs.CV↗

arXiv2021

Nonlocal effects of low-energy excitations in quantum-spin-liquid candidate Cu$_3$Zn(OH)$_6$FBr

We systematically study the low-temperature specific heats for the two-dimensional kagome antiferromagnet, Cu$_{3}$Zn(OH)$_6$FBr. The specific heat exhibits a $T^{1.7}$ dependence at low temperatures and a shoulder-like feature above it. We construct a microscopic lattice model of $Z_2$ quantum spin liquid and perform large-scale quantum Monte Carlo simulations to show that the above behaviors come from the contributions from gapped anyons and magnetic impurities. Surprisingly, we find the entropy associated with the shoulder decreases quickly with grain size $d$, although the system is paramagnetic to the lowest temperature. While this can be simply explained by a core-shell picture in that the contribution from the interior state disappears near the surface, the 5.9-nm shell width precludes any trivial explanations. Such a large length scale signifies the coherence length of the nonlocality of the quantum entangled excitations in quantum spin liquid candidate, similar to Pippard's coherence length in superconductors. Our approach therefore offers a new experimental probe of the intangible quantum state of matter with topological order.

cond-mat.str-el↗

arXiv2021

Spectral Integrals of Bernoulli Generalized Functionals

Let $\mathcal{S}\subset \mathcal{L}^2 \subset \mathcal{S}^*$ be the Gel'fand triple over the Bernoulli space, where elements of $\mathcal{S}^*$ are called Bernoulli generalized functionals. In this paper, we define integrals of Bernoulli generalized functionals with respect to a spectral measure (projection operator-valued measure) in the framework of $\mathcal{S}\subset \mathcal{L}^2 \subset \mathcal{S}^*$, and examine their fundamental properties. New notions are introduced, several results are obtained and examples are also shown.

math.FA↗

arXiv2021

Multi-Task Audio Source Separation

The audio source separation tasks, such as speech enhancement, speech separation, and music source separation, have achieved impressive performance in recent studies. The powerful modeling capabilities of deep neural networks give us hope for more challenging tasks. This paper launches a new multi-task audio source separation (MTASS) challenge to separate the speech, music, and noise signals from the monaural mixture. First, we introduce the details of this task and generate a dataset of mixtures containing speech, music, and background noises. Then, we propose an MTASS model in the complex domain to fully utilize the differences in spectral characteristics of the three audio signals. In detail, the proposed model follows a two-stage pipeline, which separates the three types of audio signals and then performs signal compensation separately. After comparing different training targets, the complex ratio mask is selected as a more suitable target for the MTASS. The experimental results also indicate that the residual signal compensation module helps to recover the signals further. The proposed model shows significant advantages in separation performance over several well-known separation models.

eess.AS↗

arXiv2021

Incorporating Multi-Target in Multi-Stage Speech Enhancement Model for Better Generalization

Recent single-channel speech enhancement methods based on deep neural networks (DNNs) have achieved remarkable results, but there are still generalization problems in real scenes. Like other data-driven methods, DNN-based speech enhancement models produce significant performance degradation on untrained data. In this study, we make full use of the contribution of multi-target joint learning to the model generalization capability, and propose a lightweight and low-computing dilated convolutional network (DCN) model for a more robust speech denoising task. Our goal is to integrate the masking target, the mapping target, and the parameters of the traditional speech enhancement estimator into a DCN model to maximize their complementary advantages. To do this, we build a multi-stage learning framework to deal with multiple targets in stages to achieve their joint learning, namely `MT-in-MS'. Our experimental results show that compared with the state-of-the-art time domain and time-frequency domain models, this proposed low-cost DCN model can achieve better generalization performance in speaker, noise, and channel mismatch cases.

eess.AS↗

arXiv2021

An Efficient Network Solver for Dynamic Simulation of Power Systems Based on Hierarchical Inverse Computation and Modification

In power system dynamic simulation, up to 90% of the computational time is devoted to solve the network equations, i.e., a set of linear equations. Traditional approaches are based on sparse LU factorization, which is inherently sequential. In this paper, an inverse-based network solution is proposed by a hierarchical method for computing and store the approximate inverse of the conductance matrix in electromagnetic transient (EMT) simulations. The proposed method can also efficiently update the inverse by modifying only local sub-matrices to reflect changes in the network, e.g., loss of a line. Experiments on a series of simplified 179-bus Western Interconnection demonstrate the advantages of the proposed methods.

eess.SY↗