arXiv subjects

Lin Xu

Publications and source records attributed to Lin Xu.

At least 55 records · Page 3Linked to original sources

arXiv2021

EMface: Detecting Hard Faces by Exploring Receptive Field Pyraminds

Scale variation is one of the most challenging problems in face detection. Modern face detectors employ feature pyramids to deal with scale variation. However, it might break the feature consistency across different scales of faces. In this paper, we propose a simple yet effective method named the receptive field pyramids (RFP) method to enhance the representation ability of feature pyramids. It can learn different receptive fields in each feature map adaptively based on the varying scales of detected faces. Empirical results on two face detection benchmark datasets, i.e., WIDER FACE and UFDD, demonstrate that our proposed method can accelerate the inference rate significantly while achieving state-of-the-art performance. The source code of our method is available at \url{https://github.com/emdata-ailab/EMface}.

cs.CV↗

arXiv2021

IDEAL: Independent Domain Embedding Augmentation Learning

Many efforts have been devoted to designing sampling, mining, and weighting strategies in high-level deep metric learning (DML) loss objectives. However, little attention has been paid to low-level but essential data transformation. In this paper, we develop a novel mechanism, the independent domain embedding augmentation learning ({IDEAL}) method. It can simultaneously learn multiple independent embedding spaces for multiple domains generated by predefined data transformations. Our IDEAL is orthogonal to existing DML techniques and can be seamlessly combined with prior DML approaches for enhanced performance. Empirical results on visual retrieval tasks demonstrate the superiority of the proposed method. For example, the IDEAL improves the performance of MS loss by a large margin, 84.5\% $\rightarrow$ 87.1\% on Cars-196, and 65.8\% $\rightarrow$ 69.5\% on CUB-200 at Recall$@1$. Our IDEAL with MS loss also achieves the new state-of-the-art performance on three image retrieval benchmarks, \ie, \emph{Cars-196}, \emph{CUB-200}, and \emph{SOP}. It outperforms the most recent DML approaches, such as Circle loss and XBM, significantly. The source code and pre-trained models of our method will be available at\emph{\url{https://github.com/emdata-ailab/IDEAL}}.

cs.CV↗

arXiv2021

Unifying Relational Sentence Generation and Retrieval for Medical Image Report Composition

Beyond generating long and topic-coherent paragraphs in traditional captioning tasks, the medical image report composition task poses more task-oriented challenges by requiring both the highly-accurate medical term diagnosis and multiple heterogeneous forms of information including impression and findings. Current methods often generate the most common sentences due to dataset bias for individual case, regardless of whether the sentences properly capture key entities and relationships. Such limitations severely hinder their applicability and generalization capability in medical report composition where the most critical sentences lie in the descriptions of abnormal diseases that are relatively rare. Moreover, some medical terms appearing in one report are often entangled with each other and co-occurred, e.g. symptoms associated with a specific disease. To enforce the semantic consistency of medical terms to be incorporated into the final reports and encourage the sentence generation for rare abnormal descriptions, we propose a novel framework that unifies template retrieval and sentence generation to handle both common and rare abnormality while ensuring the semantic-coherency among the detected medical terms. Specifically, our approach exploits hybrid-knowledge co-reasoning: i) explicit relationships among all abnormal medical terms to induce the visual attention learning and topic representation encoding for better topic-oriented symptoms descriptions; ii) adaptive generation mode that changes between the template retrieval and sentence generation according to a contextual topic encoder. Experimental results on two medical report benchmarks demonstrate the superiority of the proposed framework in terms of both human and metrics evaluation.

cs.CL↗

arXiv2020

Polarimetric SAR Image Semantic Segmentation with 3D Discrete Wavelet Transform and Markov Random Field

Polarimetric synthetic aperture radar (PolSAR) image segmentation is currently of great importance in image processing for remote sensing applications. However, it is a challenging task due to two main reasons. Firstly, the label information is difficult to acquire due to high annotation costs. Secondly, the speckle effect embedded in the PolSAR imaging process remarkably degrades the segmentation performance. To address these two issues, we present a contextual PolSAR image semantic segmentation method in this paper.With a newly defined channelwise consistent feature set as input, the three-dimensional discrete wavelet transform (3D-DWT) technique is employed to extract discriminative multi-scale features that are robust to speckle noise. Then Markov random field (MRF) is further applied to enforce label smoothness spatially during segmentation. By simultaneously utilizing 3D-DWT features and MRF priors for the first time, contextual information is fully integrated during the segmentation to ensure accurate and smooth segmentation. To demonstrate the effectiveness of the proposed method, we conduct extensive experiments on three real benchmark PolSAR image data sets. Experimental results indicate that the proposed method achieves promising segmentation accuracy and preferable spatial consistency using a minimal number of labeled pixels.

cs.CV↗

arXiv2020

Duplex Mikaelian lenses and duplex Maxwell's fish eye lenses

In this paper, we report two new kinds of absolute optical instruments that can make stigmatically images for geometric optics in two dimensional space. One is called the duplex Mikaelian lens, which is made by splicing two half Mikaelian lenses with different periods. The other is exponential conformal transformer of duplex Mikaelian lens with the ratio of different periods of its two half Mikaelian lenses a rational number, which we call duplex Maxwell's fish eye lens. Duplex Mikaelian lenses have continuous translation symmetry with arbitrary real number, while duplex Maxwell's fish eye lenses have continuous rotational symmetry from 0 to 2*Pi. Hence each duplex Maxwell's fish eye lens corresponds to a duplex Mikaelian lens. We further demonstrate the caustic effect of geometric optics in duplex Mikaelian lenses and duplex Maxwell's fish eye lenses. In addition, we investigate the Talbot effect of wave optics in the duplex Mikaelian lens based on numeric calculations. Our findings based on splicing and exponential conformal mapping enlarge the family of absolute optical instruments.

physics.optics↗

arXiv2019

A Preliminary Study on Optimal Placement of Cameras

This paper primarily focuses on figuring out the best array of cameras, or visual sensors, so that such a placement enables the maximum utilization of these visual sensors. Maximizing the utilization of these cameras can convert to another problem that is simpler for the formulation, that is, maximizing the total coverage with these cameras. To solve the problem, the coverage problem is first defined subject to the capabilities and limits of cameras. Then, poses of cameras are analyzed for the best arrangement.

cs.CV↗

arXiv2019

Detection and Classification of Breast Cancer Metastates Based on U-Net

This paper presents U-net based breast cancer metastases detection and classification in lymph nodes, as well as patient-level classification based on metastases detection. The whole pipeline can be divided into five steps: preprocessing and data argumentation, patch-based segmentation, post processing, slide-level classification, and patient-level classification. In order to reduce overfitting and speedup convergence, we applied batch normalization and dropout into U-Net. The final Kappa score reaches 0.902 on training data.

eess.IV↗

arXiv2019

DaTscan SPECT Image Classification for Parkinson's Disease

Parkinson's Disease (PD) is a neurodegenerative disease that currently does not have a cure. In order to facilitate disease management and reduce the speed of symptom progression, early diagnosis is essential. The current clinical, diagnostic approach is to have radiologists perform human visual analysis of the degeneration of dopaminergic neurons in the substantia nigra region of the brain. Clinically, dopamine levels are monitored through observing dopamine transporter (DaT) activity. One method of DaT activity analysis is performed with the injection of an Iodine-123 fluoropropyl (123I-FP-CIT) tracer combined with single photon emission computerized tomography (SPECT) imaging. The tracer illustrates the region of interest in the resulting DaTscan SPECT images. Human visual analysis is slow and vulnerable to subjectivity between radiologists, so the goal was to develop an introductory implementation of a deep convolutional neural network that can objectively and accurately classify DaTscan SPECT images as Parkinson's Disease or normal. This study illustrates the approach of using a deep convolutional neural network and evaluates its performance on DaTscan SPECT image classification. The data used in this study was obtained through a database provided by the Parkinson's Progression Markers Initiative (PPMI). The deep neural network in this study utilizes the InceptionV3 architecture, 1st runner up in the 2015 ImageNet Large Scale Visual Recognition Competition (ILSVRC), as a base model. A custom, binary classifier block was added on top of this base. In order to account for the small dataset size, a ten fold cross validation was implemented to evaluate the model's performance.

eess.IV↗

arXiv2019

HorNet: A Hierarchical Offshoot Recurrent Network for Improving Person Re-ID via Image Captioning

Person re-identification (re-ID) aims to recognize a person-of-interest across different cameras with notable appearance variance. Existing research works focused on the capability and robustness of visual representation. In this paper, instead, we propose a novel hierarchical offshoot recurrent network (HorNet) for improving person re-ID via image captioning. Image captions are semantically richer and more consistent than visual attributes, which could significantly alleviate the variance. We use the similarity preserving generative adversarial network (SPGAN) and an image captioner to fulfill domain transfer and language descriptions generation. Then the proposed HorNet can learn the visual and language representation from both the images and captions jointly, and thus enhance the performance of person re-ID. Extensive experiments are conducted on several benchmark datasets with or without image captions, i.e., CUHK03, Market-1501, and Duke-MTMC, demonstrating the superiority of the proposed method. Our method can generate and extract meaningful image captions while achieving state-of-the-art performance.

cs.CV↗

arXiv2019

Anisotropy engineering edge magnetism in zigzag honeycomb nanoribbons

It have been demonstrated that the zigzag honeycomb nanoribbons exhibit an intriguing edge magnetism. Here the effect of the anisotropy on the edge magnetism in zigzag honeycomb nanoribbons is investigated by using two kinds of large-scale quantum Monte Carlo simulations. The anisotropy in zigzag honeycomb nanoribbons is characterized by the ratios of nearest-neighboring hopping integrals $t_{1}$ in one direction and $t_{2}$ in another direction. Considering the electron-electron correlation, it is shown that the edge ferromagnetism could be enhanced greatly as $t_{2}/|t_{1}|$ increase from $1$ to $3$, which not only presenting the avenue for the control of this magnetism, but also being useful for exploring further novel magnetism in new nano-scale materials.

cond-mat.str-el↗

arXiv2019

Learning with Batch-wise Optimal Transport Loss for 3D Shape Recognition

Deep metric learning is essential for visual recognition. The widely used pair-wise (or triplet) based loss objectives cannot make full use of semantical information in training samples or give enough attention to those hard samples during optimization. Thus, they often suffer from a slow convergence rate and inferior performance. In this paper, we show how to learn an importance-driven distance metric via optimal transport programming from batches of samples. It can automatically emphasize hard examples and lead to significant improvements in convergence. We propose a new batch-wise optimal transport loss and combine it in an end-to-end deep metric learning manner. We use it to learn the distance metric and deep feature representation jointly for recognition. Empirical results on visual retrieval and classification tasks with six benchmark datasets, i.e., MNIST, CIFAR10, SHREC13, SHREC14, ModelNet10, and ModelNet40, demonstrate the superiority of the proposed method. It can accelerate the convergence rate significantly while achieving a state-of-the-art recognition performance. For example, in 3D shape recognition experiments, we show that our method can achieve better recognition performance within only 5 epochs than what can be obtained by mainstream 3D shape recognition approaches after 200 epochs.

cs.CV↗

arXiv2019

End-to-End Knowledge-Routed Relational Dialogue System for Automatic Diagnosis

Beyond current conversational chatbots or task-oriented dialogue systems that have attracted increasing attention, we move forward to develop a dialogue system for automatic medical diagnosis that converses with patients to collect additional symptoms beyond their self-reports and automatically makes a diagnosis. Besides the challenges for conversational dialogue systems (e.g. topic transition coherency and question understanding), automatic medical diagnosis further poses more critical requirements for the dialogue rationality in the context of medical knowledge and symptom-disease relations. Existing dialogue systems (Madotto, Wu, and Fung 2018; Wei et al. 2018; Li et al. 2017) mostly rely on data-driven learning and cannot be able to encode extra expert knowledge graph. In this work, we propose an End-to-End Knowledge-routed Relational Dialogue System (KR-DS) that seamlessly incorporates rich medical knowledge graph into the topic transition in dialogue management, and makes it cooperative with natural language understanding and natural language generation. A novel Knowledge-routed Deep Q-network (KR-DQN) is introduced to manage topic transitions, which integrates a relational refinement branch for encoding relations among different symptoms and symptom-disease pairs, and a knowledge-routed graph branch for topic decision-making. Extensive experiments on a public medical dialogue dataset show our KR-DS significantly beats state-of-the-art methods (by more than 8% in diagnosis accuracy). We further show the superiority of our KR-DS on a newly collected medical dialogue system dataset, which is more challenging retaining original self-reports and conversational data between patients and doctors.

cs.CL↗

arXiv2018

Semiparametric Mixture Regression with Unspecified Error Distributions

In fitting a mixture of linear regression models, normal assumption is traditionally used to model the error and then regression parameters are estimated by the maximum likelihood estimators (MLE). This procedure is not valid if the normal assumption is violated. To relax the normal assumption on the error distribution hence reduce the modeling bias, we propose semiparametric mixture of linear regression models with unspecified error distributions. We establish a more general identifiability result under weaker conditions than existing results, construct a class of new estimators, and establish their asymptotic properties. These asymptotic results also apply to many existing semiparametric mixture regression estimators whose asymptotic properties have remained unknown due to the inherent difficulties in obtaining them. Using simulation studies, we demonstrate the superiority of the proposed estimators over the MLE when the normal error assumption is violated and the comparability when the error is normal. Analysis of a newly collected Equine Infectious Anemia Virus data in 2017 is employed to illustrate the usefulness of the new estimator.

stat.ME↗

arXiv2018

Transformation devices with optical nihility media and reduced realizations

Starting from optical nihility media (ONM), we design several intriguing devices with transformation optics method in two dimensions, such as a wave splitter, a concave lens, a field rotator, a concentrator and an invisibility cloak. The extreme anisotropic property of ONM hinders the fabrication of these devices, which could be effectively realized by simplified materials with Fabry-Pérot resonances (FPs) at discrete frequencies. Moreover, we propose a reduced version of simplified materials with FPs to construct a concentrator and a rotator, which is feasible in experimental fabrications. The simulations of total scattering cross sections confirm their functionalities.

physics.optics↗

arXiv2018

Three-dimensional photonic Dirac points stabilized by point group symmetry

We discover a pair of stable 3D Dirac points, 3D photonic analog of graphene, in all-dielectric photonic crystals using structures commensurate with nano-fabrication for visible-frequency photonic applications. The Dirac points carry nontrivial $Z_2$ topology and emerge for a large range of material parameters in hollow cylinder hexagonal photonic crystals. From Kramers theorem and group theory, we find that only the $C_6$ symmetry lead to point group symmetry stabilized Dirac points in 3D all-dielectric photonic crystals. {The Dirac points are characterized using ${\vec k}\cdot{\vec P}$ theory for photonic bands in combination with symmetry analysis. Breaking inversion symmetry splites the Dirac points into Weyl points. The physical properties and experimental consequences of Dirac points are also studied. The Dirac points are found to be robust against parameter tuning and weak disorders.

cond-mat.mtrl-sci↗

arXiv2018

Geodesic conformal transformation optics: manipulating light with continuous refractive index profile

Conformal transformation optics provides a simple scheme for manipulating light rays with inhomogeneous isotropic dielectrics. However, there is usually discontinuity for refractive index profile at branch cuts of different virtual Riemann sheets, hence compromising the functionalities. To deal with that, we present a special method for conformal transformation optics based on the concept of geodesic lens. The requirement is a continuous refractive index profile of dielectrics, which shows almost perfect performance of designed devices. We demonstrate such a proposal by achieving conformal transparency and reflection. We can further achieve conformal invisible cloaks by two techniques with perfect electromagnetic conductors. The geodesic concept may also find applications in other waves that obey the Helmholtz equation in two dimensions.

physics.optics↗

arXiv2018

Light rays and waves on geodesic lenses

Starting from well-known absolute instruments for perfect imaging, we introduce a type of rotational-symmetrical compact closed manifolds, namely geodesic lenses. We demonstrate that light rays confined on geodesic lenses are closed trajectories. While for optical waves, the spectrum of geodesic lens is (at least approximately) degenerate and equidistant with numerical methods. Based on this property, we show a periodical evolution of optical waves and quantum waves on geodesic lenses. Moreover, we fabricate two geodesic lenses in sub-micrometer scale, where curved light rays are observed with high accurate precision. Our results may offer a new platform to investigate light propagation on curved surfaces.

physics.optics↗

arXiv2017

Hyperspectral Image Classification with Markov Random Fields and a Convolutional Neural Network

This paper presents a new supervised classification algorithm for remotely sensed hyperspectral image (HSI) which integrates spectral and spatial information in a unified Bayesian framework. First, we formulate the HSI classification problem from a Bayesian perspective. Then, we adopt a convolutional neural network (CNN) to learn the posterior class distributions using a patch-wise training strategy to better use the spatial information. Next, spatial information is further considered by placing a spatial smoothness prior on the labels. Finally, we iteratively update the CNN parameters using stochastic gradient decent (SGD) and update the class labels of all pixel vectors using an alpha-expansion min-cut-based algorithm. Compared with other state-of-the-art methods, the proposed classification method achieves better performance on one synthetic dataset and two benchmark HSI datasets in a number of experimental settings.

cs.CV↗