arXiv subjects

Yu Zhang

Publications and source records attributed to Yu Zhang.

At least 1,351 records · Page 75Linked to original sources

arXiv2020

A Study on Evaluation Standard for Automatic Crack Detection Regard the Random Fractal

A reasonable evaluation standard underlies construction of effective deep learning models. However, we find in experiments that the automatic crack detectors based on deep learning are obviously underestimated by the widely used mean Average Precision (mAP) standard. This paper presents a study on the evaluation standard. It is clarified that the random fractal of crack disables the mAP standard, because the strict box matching in mAP calculation is unreasonable for the fractal feature. As a solution, a fractal-available evaluation standard named CovEval is proposed to correct the underestimation in crack detection. In CovEval, a different matching process based on the idea of covering box matching is adopted for this issue. In detail, Cover Area rate (CAr) is designed as a covering overlap, and a multi-match strategy is employed to release the one-to-one matching restriction in mAP. Extended Recall (XR), Extended Precision (XP) and Extended F-score (Fext) are defined for scoring the crack detectors. In experiments using several common frameworks for object detection, models get much higher scores in crack detection according to CovEval, which matches better with the visual performance. Moreover, based on faster R-CNN framework, we present a case study to optimize a crack detector based on CovEval standard. Recall (XR) of our best model achieves an industrial-level at 95.8, which implies that with reasonable standard for evaluation, the methods for object detection are with great potential for automatic industrial inspection.

cs.CV↗

arXiv2020

RGB-D SLAM in Dynamic Environments Using Point Correlations

In this paper, a simultaneous localization and mapping (SLAM) method that eliminates the influence of moving objects in dynamic environments is proposed. This method utilizes the correlation between map points to separate points that are part of the static scene and points that are part of different moving objects into different groups. A sparse graph is first created using Delaunay triangulation from all map points. In this graph, the vertices represent map points, and each edge represents the correlation between adjacent points. If the relative position between two points remains consistent over time, there is correlation between them, and they are considered to be moving together rigidly. If not, they are considered to have no correlation and to be in separate groups. After the edges between the uncorrelated points are removed during point-correlation optimization, the remaining graph separates the map points of the moving objects from the map points of the static scene. The largest group is assumed to be the group of reliable static map points. Finally, motion estimation is performed using only these points. The proposed method was implemented for RGB-D sensors, evaluated with a public RGB-D benchmark, and tested in several additional challenging environments. The experimental results demonstrate that robust and accurate performance can be achieved by the proposed SLAM method in both slightly and highly dynamic environments. Compared with other state-of-the-art methods, the proposed method can provide competitive accuracy with good real-time performance.

cs.CV↗

arXiv2020

Deep Image Clustering with Category-Style Representation

Deep clustering which adopts deep neural networks to obtain optimal representations for clustering has been widely studied recently. In this paper, we propose a novel deep image clustering framework to learn a category-style latent representation in which the category information is disentangled from image style and can be directly used as the cluster assignment. To achieve this goal, mutual information maximization is applied to embed relevant information in the latent representation. Moreover, augmentation-invariant loss is employed to disentangle the representation into category part and style part. Last but not least, a prior distribution is imposed on the latent representation to ensure the elements of the category vector can be used as the probabilities over clusters. Comprehensive experiments demonstrate that the proposed approach outperforms state-of-the-art methods significantly on five public datasets.

cs.CV↗

arXiv2020

Hierarchical Topic Mining via Joint Spherical Tree and Text Embedding

Mining a set of meaningful topics organized into a hierarchy is intuitively appealing since topic correlations are ubiquitous in massive text corpora. To account for potential hierarchical topic structures, hierarchical topic models generalize flat topic models by incorporating latent topic hierarchies into their generative modeling process. However, due to their purely unsupervised nature, the learned topic hierarchy often deviates from users' particular needs or interests. To guide the hierarchical topic discovery process with minimal user supervision, we propose a new task, Hierarchical Topic Mining, which takes a category tree described by category names only, and aims to mine a set of representative terms for each category from a text corpus to help a user comprehend his/her interested topics. We develop a novel joint tree and text embedding method along with a principled optimization procedure that allows simultaneous modeling of the category tree structure and the corpus generative process in the spherical space for effective category-representative term discovery. Our comprehensive experiments show that our model, named JoSH, mines a high-quality set of hierarchical topics with high efficiency and benefits weakly-supervised hierarchical text classification tasks.

cs.CL↗

arXiv2020

Centrality selection effect on higher-order cumulants of net-proton multiplicity distributions in relativistic heavy-ion collisions

We studied the centrality selection effect on cumulants (up to fourth order) and the cumulants ratios of net-proton multiplicity distributions in Au+Au collisions at $\sqrt{s_{\mathrm{NN}}}$ = 7.7, 19.6 and 200 GeV from UrQMD model. The net-proton cumulants are calculated with collision centralities by using charged particle multiplicity from different pesudorapidity ($η$) region. By comparing the results from various collision centralities, we found that the autocorrelation effects are not significant in the results with collision centralities "refmult-3" and "refmult-2", which are using mid-rapidity charged particles but excluding (anti-)protons and analysis region, respectively. Furthermore, due to the contributions of spectator protons, we observed poor centrality resolution when using charged particles at forward $η$ region at low energies. This work can serve as a baseline for centrality selection of future fluctuations analysis in relativistic heavy-ion collisions.

nucl-ex↗

arXiv2020

ASVspoof 2019: A large-scale public database of synthesized, converted and replayed speech

Automatic speaker verification (ASV) is one of the most natural and convenient means of biometric person recognition. Unfortunately, just like all other biometric systems, ASV is vulnerable to spoofing, also referred to as "presentation attacks." These vulnerabilities are generally unacceptable and call for spoofing countermeasures or "presentation attack detection" systems. In addition to impersonation, ASV systems are vulnerable to replay, speech synthesis, and voice conversion attacks. The ASVspoof 2019 edition is the first to consider all three spoofing attack types within a single challenge. While they originate from the same source database and same underlying protocol, they are explored in two specific use case scenarios. Spoofing attacks within a logical access (LA) scenario are generated with the latest speech synthesis and voice conversion technologies, including state-of-the-art neural acoustic and waveform model techniques. Replay spoofing attacks within a physical access (PA) scenario are generated through carefully controlled simulations that support much more revealing analysis than possible previously. Also new to the 2019 edition is the use of the tandem detection cost function metric, which reflects the impact of spoofing and countermeasures on the reliability of a fixed ASV system. This paper describes the database design, protocol, spoofing attack implementations, and baseline ASV and countermeasure results. It also describes a human assessment on spoofed data in logical access. It was demonstrated that the spoofing data in the ASVspoof 2019 database have varied degrees of perceived quality and similarity to the target speakers, including spoofed data that cannot be differentiated from bona-fide utterances even by human subjects.

eess.AS↗

arXiv2020

M2Net: Multi-modal Multi-channel Network for Overall Survival Time Prediction of Brain Tumor Patients

Early and accurate prediction of overall survival (OS) time can help to obtain better treatment planning for brain tumor patients. Although many OS time prediction methods have been developed and obtain promising results, there are still several issues. First, conventional prediction methods rely on radiomic features at the local lesion area of a magnetic resonance (MR) volume, which may not represent the full image or model complex tumor patterns. Second, different types of scanners (i.e., multi-modal data) are sensitive to different brain regions, which makes it challenging to effectively exploit the complementary information across multiple modalities and also preserve the modality-specific properties. Third, existing methods focus on prediction models, ignoring complex data-to-label relationships. To address the above issues, we propose an end-to-end OS time prediction model; namely, Multi-modal Multi-channel Network (M2Net). Specifically, we first project the 3D MR volume onto 2D images in different directions, which reduces computational costs, while preserving important information and enabling pre-trained models to be transferred from other tasks. Then, we use a modality-specific network to extract implicit and high-level features from different MR scans. A multi-modal shared network is built to fuse these features using a bilinear pooling model, exploiting their correlations to provide complementary information. Finally, we integrate the outputs from each modality-specific network and the multi-modal shared network to generate the final prediction result. Experimental results demonstrate the superiority of our M2Net model over other methods.

eess.IV↗

arXiv2020

Multi-Site Infant Brain Segmentation Algorithms: The iSeg-2019 Challenge

To better understand early brain growth patterns in health and disorder, it is critical to accurately segment infant brain magnetic resonance (MR) images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF). Deep learning-based methods have achieved state-of-the-art performance; however, one of major limitations is that the learning-based methods may suffer from the multi-site issue, that is, the models trained on a dataset from one site may not be applicable to the datasets acquired from other sites with different imaging protocols/scanners. To promote methodological development in the community, iSeg-2019 challenge (http://iseg2019.web.unc.edu) provides a set of 6-month infant subjects from multiple sites with different protocols/scanners for the participating methods. Training/validation subjects are from UNC (MAP) and testing subjects are from UNC/UMN (BCP), Stanford University, and Emory University. By the time of writing, there are 30 automatic segmentation methods participating in iSeg-2019. We review the 8 top-ranked teams by detailing their pipelines/implementations, presenting experimental results and evaluating performance in terms of the whole brain, regions of interest, and gyral landmark curves. We also discuss their limitations and possible future directions for the multi-site issue. We hope that the multi-site dataset in iSeg-2019 and this review article will attract more researchers on the multi-site issue.

eess.IV↗

arXiv2020

Tunable lattice reconstruction and bandwidth of flat bands in magic-angle twisted bilayer graphene

The interplay between interlayer van der Waals interaction and intralayer lattice distortion can lead to structural reconstruction in slightly twisted bilayer graphene (TBG) with the twist angle being smaller than a characteristic angle θc. Experimentally, the θc is demonstrated to be very close to the magic angle (θ ~ 1.05°). In this work, we address the transition between reconstructed and unreconstructed structures of the TBG across the magic angle by using scanning tunnelling microscopy (STM). Our experiment demonstrates that both the two structures are stable in the TBG around the magic angle. By applying a STM tip pulse, we show that the two structures can be switched to each other and the bandwidth of the flat bands, which plays a vital role in the emergent strongly correlated states in the magic-angle TBG, can be tuned. The observed tunable lattice reconstruction and bandwidth of the flat bands provide an extra control knob to manipulate the exotic electronic states of the TBG near the magic angle.

cond-mat.mes-hall↗

arXiv2020

Robust two-dimensional ice on graphene built from finite-length water molecular chains

Interfacial ice on graphene has attracted much attention because it is a model system to study two-dimensional (2D) ice structures on chemically inert substrates. While water-graphene interaction was usually assumed to be negligible, the structures of the 2D ice are believed to be not appreciably perturbed by the graphene substrate. Here we report atomic-resolved characterizations of an exotic 2D ice structure on graphene built from water molecular chains with finite lengths. Our experiments demonstrated that the water molecular chains are exactly orientated along zigzag directions of the graphene substrate, which evidences an anomalously strong interlayer interaction between the 2D ice and the graphene substrate. Moreover, the length of the water molecular chains closely links to the number of graphene layers, indicating layer-number-dependent water-graphene interfacial interactions. Our work highlights the important role of the 2D ice structures on the water-graphene interfacial interactions.

cond-mat.mtrl-sci↗

arXiv2020

Achieving Multi-Tasking Robots in Multi-Robot Tasks

One simplifying assumption made in distributed robot systems is that the robots are single-tasking: each robot operates on a single task at any time. While such a sanguine assumption is innocent to make in situations with sufficient resources so that the robots can operate independently, it becomes impractical when they must share their capabilities. In this paper, we consider multi-tasking robots with multi-robot tasks. Given a set of tasks, each achievable by a coalition of robots, our approach allows the coalitions to overlap and task synergies to be exploited by reasoning about the physical constraints that can be synergistically satisfied for achieving the tasks. The key contribution of this work is a general and flexible framework to achieve this ability for multi-robot systems in resource-constrained situations to extend their capabilities. The proposed approach is built on the information invariant theory, which specifies the interactions between information requirements. In our work, we map physical constraints to information requirements, thereby allowing task synergies to be identified via the information invariant framework. We show that our algorithm is sound and complete under a problem setting with multi-tasking robots. Simulation results show its effectiveness under resource-constrained situations and in handling challenging situations in a multi-UAV simulator.

cs.RO↗

arXiv2020

Allocation of Multi-Robot Tasks with Task Variants

Task allocation has been a well studied problem. In most prior problem formulations, it is assumed that each task is associated with a unique set of resource requirements. In the scope of multi-robot task allocation problem, these requirements can be satisfied by a coalition of robots. In this paper, we introduce a more general formulation of multi-robot task allocation problem that allows more than one option for specifying the set of task requirements--satisfying any one of the options will satisfy the task. We referred to this new problem as the multi-robot task allocation problem with task variants. First, we theoretically show that this extension fortunately does not impact the complexity class, which is still NP-complete. For solution methods, we adapt two previous greedy methods for the task allocation problem without task variants to solve this new problem and analyze their effectiveness. In particular, we "flatten" the new problem to the problem without task variants, modify the previous methods to solve the flattened problem, and prove that the bounds still hold. Finally, we thoroughly evaluate these two methods along with a random baseline to demonstrate their efficacy for the new problem.

cs.AI↗

arXiv2020

Efficient Second-Order TreeCRF for Neural Dependency Parsing

In the deep learning (DL) era, parsing models are extremely simplified with little hurt on performance, thanks to the remarkable capability of multi-layer BiLSTMs in context representation. As the most popular graph-based dependency parser due to its high efficiency and performance, the biaffine parser directly scores single dependencies under the arc-factorization assumption, and adopts a very simple local token-wise cross-entropy training loss. This paper for the first time presents a second-order TreeCRF extension to the biaffine parser. For a long time, the complexity and inefficiency of the inside-outside algorithm hinder the popularity of TreeCRF. To address this issue, we propose an effective way to batchify the inside and Viterbi algorithms for direct large matrix operation on GPUs, and to avoid the complex outside algorithm via efficient back-propagation. Experiments and analysis on 27 datasets from 13 languages clearly show that techniques developed before the DL era, such as structural learning (global TreeCRF loss) and high-order modeling are still useful, and can further boost parsing performance over the state-of-the-art biaffine parser, especially for partially annotated training data. We release our code at https://github.com/yzhangcs/crfpar.

cs.CL↗

arXiv2020

Offline Handwritten Chinese Text Recognition with Convolutional Neural Networks

Deep learning based methods have been dominating the text recognition tasks in different and multilingual scenarios. The offline handwritten Chinese text recognition (HCTR) is one of the most challenging tasks because it involves thousands of characters, variant writing styles and complex data collection process. Recently, the recurrent-free architectures for text recognition appears to be competitive as its highly parallelism and comparable results. In this paper, we build the models using only the convolutional neural networks and use CTC as the loss function. To reduce the overfitting, we apply dropout after each max-pooling layer and with extreme high rate on the last one before the linear layer. The CASIA-HWDB database is selected to tune and evaluate the proposed models. With the existing text samples as templates, we randomly choose isolated character samples to synthesis more text samples for training. We finally achieve 6.81% character error rate (CER) on the ICDAR 2013 competition set, which is the best published result without language model correction.

cs.CV↗

arXiv2020

Neural Networks Based Beam Codebooks: Learning mmWave Massive MIMO Beams that Adapt to Deployment and Hardware

Millimeter wave (mmWave) and massive MIMO systems are intrinsic components of 5G and beyond. These systems rely on using beamforming codebooks for both initial access and data transmission. Current beam codebooks, however, generally consist of a large number of narrow beams that scan all possible directions, even if these directions are never used. This leads to very large training overhead. Further, these codebooks do not normally account for the hardware impairments or the possible non-uniform array geometries, and their calibration is an expensive process. To overcome these limitations, this paper develops an efficient online machine learning framework that learns how to adapt the codebook beam patterns to the specific deployment, surrounding environment, user distribution, and hardware characteristics. This is done by designing a novel complex-valued neural network architecture in which the neuron weights directly model the beamforming weights of the analog phase shifters, accounting for the key hardware constraints such as the constant-modulus and quantized-angles. This model learns the codebook beams through online and self-supervised training avoiding the need for explicit channel state information. This respects the practical situations where the channel is either unavailable, imperfect, or hard to obtain, especially in the presence of hardware impairments. Simulation results highlight the capability of the proposed solution in learning environment and hardware aware beam codebooks, which can significantly reduce the training overhead, enhance the achievable data rates, and improve the robustness against possible hardware impairments.

eess.SP↗

arXiv2020

Brain2Object: Printing Your Mind from Brain Signals with Spatial Correlation Embedding

Electroencephalography (EEG) signals are known to manifest differential patterns when individuals visually concentrate on different objects. In this work, we present an end-to-end digital fabrication system, Brain2Object, to print the 3D object that an individual is observing by decoding visually-evoked brain signals. We propose a unified training framework that combines multi-class Common Spatial Pattern and Convolutional Neural Networks to support the backend computation. We learn the dynamical graph representations of brain signals to accurately capture the structural information among EEG channels. A user-friendly interface is developed as the system front end. Brain2Object presents a streamlined end-to-end workflow that can serve as a template for deeper integration of BCI technologies to assist with our routine activities. The proposed system is evaluated extensively using offline experiments and through an online demonstrator. The experimental results show that our approach can achieve the recognition accuracy of 92.58% on a benchmark dataset and 75.23% on a locally collected dataset. Moreover, our method consistently outperforms a wide range of baseline and state-of-the-art approaches. The proof-of-concept corroborates the practicality of our approach and illustrates the ease with which such a system could be deployed.

cs.HC↗

arXiv2020

Momentum Contrastive Learning for Few-Shot COVID-19 Diagnosis from Chest CT Images

The current pandemic, caused by the outbreak of a novel coronavirus (COVID-19) in December 2019, has led to a global emergency that has significantly impacted economies, healthcare systems and personal wellbeing all around the world. Controlling the rapidly evolving disease requires highly sensitive and specific diagnostics. While real-time RT-PCR is the most commonly used, these can take up to 8 hours, and require significant effort from healthcare professionals. As such, there is a critical need for a quick and automatic diagnostic system. Diagnosis from chest CT images is a promising direction. However, current studies are limited by the lack of sufficient training samples, as acquiring annotated CT images is time-consuming. To this end, we propose a new deep learning algorithm for the automated diagnosis of COVID-19, which only requires a few samples for training. Specifically, we use contrastive learning to train an encoder which can capture expressive feature representations on large and publicly available lung datasets and adopt the prototypical network for classification. We validate the efficacy of the proposed model in comparison with other competing methods on two publicly available and annotated COVID-19 CT datasets. Our results demonstrate the superior performance of our model for the accurate diagnosis of COVID-19 based on chest CT images.

eess.IV↗

arXiv2020

Distant Transfer Learning via Deep Random Walk

Transfer learning, which is to improve the learning performance in the target domain by leveraging useful knowledge from the source domain, often requires that those two domains are very close, which limits its application scope. Recently, distant transfer learning has been studied to transfer knowledge between two distant or even totally unrelated domains via auxiliary domains that are usually unlabeled as a bridge in the spirit of human transitive inference that it is possible to connect two completely unrelated concepts together through gradual knowledge transfer. In this paper, we study distant transfer learning by proposing a DeEp Random Walk basEd distaNt Transfer (DERWENT) method. Different from existing distant transfer learning models that implicitly identify the path of knowledge transfer between the source and target instances through auxiliary instances, the proposed DERWENT model can explicitly learn such paths via the deep random walk technique. Specifically, based on sequences identified by the random walk technique on a data graph where source and target data have no direct edges, the proposed DERWENT model enforces adjacent data points in a squence to be similar, makes the ending data point be represented by other data points in the same sequence, and considers weighted training losses of source data. Empirical studies on several benchmark datasets demonstrate that the proposed DERWENT algorithm yields the state-of-the-art performance.

cs.LG↗