arXiv subjects

Bin Yang

Publications and source records attributed to Bin Yang.

At least 379 records · Page 21Linked to original sources

arXiv2019

Deep One-Class Classification Using Intra-Class Splitting

This paper introduces a generic method which enables to use conventional deep neural networks as end-to-end one-class classifiers. The method is based on splitting given data from one class into two subsets. In one-class classification, only samples of one normal class are available for training. During inference, a closed and tight decision boundary around the training samples is sought which conventional binary or multi-class neural networks are not able to provide. By splitting data into typical and atypical normal subsets, the proposed method can use a binary loss and defines an auxiliary subnetwork for distance constraints in the latent space. Various experiments on three well-known image datasets showed the effectiveness of the proposed method which outperformed seven baselines and had a better or comparable performance to the state-of-the-art.

cs.LG↗

arXiv2019

Enhancing Flood Impact Analysis using Interactive Retrieval of Social Media Images

The analysis of natural disasters such as floods in a timely manner often suffers from limited data due to a coarse distribution of sensors or sensor failures. This limitation could be alleviated by leveraging information contained in images of the event posted on social media platforms, so-called "Volunteered Geographic Information (VGI)". To save the analyst from the need to inspect all images posted online manually, we propose to use content-based image retrieval with the possibility of relevance feedback for retrieving only relevant images of the event to be analyzed. To evaluate this approach, we introduce a new dataset of 3,710 flood images, annotated by domain experts regarding their relevance with respect to three tasks (determining the flooded area, inundation depth, water pollution). We compare several image features and relevance feedback methods on that dataset, mixed with 97,085 distractor images, and are able to improve the precision among the top 100 retrieval results from 55% with the baseline retrieval to 87% after 5 rounds of feedback.

cs.IR↗

arXiv2019

Towards Adversarial Denoising of Radar Micro-Doppler Signatures

Generative Adversarial Networks (GANs) are considered the state-of-the-art in the field of image generation. They learn the joint distribution of the training data and attempt to generate new data samples in high dimensional space following the same distribution as the input. Recent improvements in GANs opened the field to many other computer vision applications based on improving and changing the characteristics of the input image to follow some given training requirements. In this paper, we propose a novel technique for the denoising and reconstruction of the micro-Doppler ($\boldsymbolμ$-D) spectra of walking humans based on GANs. Two sets of experiments were collected on 22 subjects walking on a treadmill at an intermediate velocity using a \unit[25]{GHz} CW radar. In one set, a clean $\boldsymbolμ$-D spectrum is collected for each subject by placing the radar at a close distance to the subject. In the other set, variations are introduced in the experiment setup to introduce different noise and clutter effects on the spectrum by changing the distance and placing reflective objects between the radar and the target. Synthetic paired noisy and noise-free spectra were used for training, while validation was carried out on the real noisy measured data. Finally, qualitative and quantitative comparison with other classical radar denoising approaches in the literature demonstrated the proposed GANs framework is better and more robust to different noise levels.

cs.CV↗

arXiv2019

Radiative transitions and magnetic moments of the charmed and bottom vector mesons in chiral perturbation theory

In this work, we systematically study the radiative decays and magnetic moments of the charmed and bottom vector mesons with chiral perturbation theory up to one-loop level. We present the results in SU(2) and SU(3) cases with the mass splitting in loop diagrams kept and unkept, respectively. The obtained decay rates for $D^\ast$ and $B^\ast$ mesons in SU(3) case with the mass splitting kept are: $Γ_{\bar{D}^{\ast 0}\to \bar{D}^0γ}=16.2^{+6.5}_{-6.0}$ keV, $Γ_{D^{\ast-}\to D^-γ}=0.73^{+0.7}_{-0.3}$ keV, $Γ_{D_s^{\ast-}\to D_s^-γ}= 0.32^{+0.3}_{-0.3}$ keV, and $Γ_{B^{\ast+}\to B^+γ}=0.58^{+0.2}_{-0.2}$ keV, $Γ_{B^{\ast0}\to B^0γ}=0.23^{+0.06}_{-0.06}$ keV, $Γ_{B_s^{\ast0}\to B_s^0γ}=0.04^{+0.03}_{-0.03}$ keV. The decay width for $D^{\ast-}\to D^-γ$ is consistent with the experimental measurement. As a byproduct, the full widths of $\bar{D}^{\ast0}$ and $D_s^{\ast-}$ are $Γ_{\mathrm{tot}}(\bar{D}^{\ast0})\simeq77.7^{+26.7}_{-20.5}~\mathrm{keV}$ and $ Γ_{\mathrm{tot}}(D_s^{\ast-})\simeq0.62^{+0.45}_{-0.50}~\mathrm{keV}$, respectively. We also calculate the magnetic moments of the heavy vector mesons. The analytical chiral expressions derived in our work shall be helpful for the extrapolations of lattice QCD simulations in the coming future.

hep-ph↗

arXiv2019

Eigenfunction Behavior and Adaptive Finite Element Approximations of Nonlinear Eigenvalue Problems in Quantum Physics

In this paper, we investigate a class of nonlinear eigenvalue problems resulting from quantum physics. We first prove that the eigenfunction cannot be a polynomial on any open set, which may be reviewed as a refinement of the classic unique continuation property. Then we apply the non-polynomial behavior of eigenfunction to show that the adaptive finite element approximations are convergent even if the initial mesh is not fine enough. We finally remark that similar arguments can be applied to a class of linear eigenvalue problems that improve the relevant existing result.

math.NA↗

arXiv2019

PathRank: A Multi-Task Learning Framework to Rank Paths in Spatial Networks

Modern navigation services often provide multiple paths connecting the same source and destination for users to select. Hence, ranking such paths becomes increasingly important, which directly affects the service quality. We present PathRank, a data-driven framework for ranking paths based on historical trajectories using multi-task learning. If a trajectory used path P from source s to destination d, PathRank considers this as an evidence that P is preferred over all other paths from s to d. Thus, a path that is similar to P should have a larger ranking score than a path that is dissimilar to P. Based on this intuition, PathRank models path ranking as a regression problem, where each path is associated with a ranking score. To enable PathRank, we first propose an effective method to generate a compact set of training data: for each trajectory, we generate a small set of diversified paths. Next, we propose a multi-task learning framework to solve the regression problem. In particular, a spatial network embedding is proposed to embed each vertex to a feature vector by considering both road network topology and spatial properties, such as distances and travel times. Since a path is represented by a sequence of vertices, which is now a sequence of feature vectors after embedding, recurrent neural network is applied to model the sequence. The objective function is designed to consider errors on both ranking scores and spatial properties, making the framework a multi-task learning framework. Empirical studies on a substantial trajectory data set offer insight into the designed properties of the proposed framework and indicating that it is effective and practical.

cs.LG↗

arXiv2019

An Adversarial Super-Resolution Remedy for Radar Design Trade-offs

Radar is of vital importance in many fields, such as autonomous driving, safety and surveillance applications. However, it suffers from stringent constraints on its design parametrization leading to multiple trade-offs. For example, the bandwidth in FMCW radars is inversely proportional with both the maximum unambiguous range and range resolution. In this work, we introduce a new method for circumventing radar design trade-offs. We propose the use of recent advances in computer vision, more specifically generative adversarial networks (GANs), to enhance low-resolution radar acquisitions into higher resolution counterparts while maintaining the advantages of the low-resolution parametrization. The capability of the proposed method was evaluated on the velocity resolution and range-azimuth trade-offs in micro-Doppler signatures and FMCW uniform linear array (ULA) radars, respectively.

cs.CV↗

arXiv2019

Possible Molecular States Composed of Doubly Charmed Baryons with Couple-channel Effect

We systematically investigate the possible molecular states composed of (1) two spin-$\frac{3}{2}$ doubly charmed baryons, and (2) a pair of spin-$\frac{3}{2}$ and spin-$\frac{1}{2}$ doubly charmed baryons. The one-boson-exchange (OBE) model is used to describe the potential between two baryons. The channel mixing effect is considered for the systems with the same quantum number $(I(J^P))$ but different total spin ($S$) and orbital angular momenta ($L$). We also study the channel mixing effect among the systems composed of various doubly charmed baryons if they have the same quantum number. Many of the systems are good candidates of molecular states.

hep-ph↗

arXiv2019

Localizing Catastrophic Forgetting in Neural Networks

Artificial neural networks (ANNs) suffer from catastrophic forgetting when trained on a sequence of tasks. While this phenomenon was studied in the past, there is only very limited recent research on this phenomenon. We propose a method for determining the contribution of individual parameters in an ANN to catastrophic forgetting. The method is used to analyze an ANNs response to three different continual learning scenarios.

cs.LG↗

arXiv2019

2-bit Model Compression of Deep Convolutional Neural Network on ASIC Engine for Image Retrieval

Image retrieval utilizes image descriptors to retrieve the most similar images to a given query image. Convolutional neural network (CNN) is becoming the dominant approach to extract image descriptors for image retrieval. For low-power hardware implementation of image retrieval, the drawback of CNN-based feature descriptor is that it requires hundreds of megabytes of storage. To address this problem, this paper applies deep model quantization and compression to CNN in ASIC chip for image retrieval. It is demonstrated that the CNN-based features descriptor can be extracted using as few as 2-bit weights quantization to deliver a similar performance as floating-point model for image retrieval. In addition, to implement CNN in ASIC, especially for large scale images, the limited buffer size of chips should be considered. To retrieve large scale images, we propose an improved pooling strategy, region nested invariance pooling (RNIP), which uses cropped sub-images for CNN. Testing results on chip show that integrating RNIP with the proposed 2-bit CNN model compression approach is capable of retrieving large scale images.

cs.AI↗

arXiv2019

Learning to Reweight Examples for Robust Deep Learning

Deep neural networks have been shown to be very powerful modeling tools for many supervised learning tasks involving complex input patterns. However, they can also easily overfit to training set biases and label noises. In addition to various regularizers, example reweighting algorithms are popular solutions to these problems, but they require careful tuning of additional hyperparameters, such as example mining schedules and regularization hyperparameters. In contrast to past reweighting methods, which typically consist of functions of the cost value of each example, in this work we propose a novel meta-learning algorithm that learns to assign weights to training examples based on their gradient directions. To determine the example weights, our method performs a meta gradient descent step on the current mini-batch example weights (which are initialized from zero) to minimize the loss on a clean unbiased validation set. Our proposed method can be easily implemented on any type of deep network, does not require any additional hyperparameter tuning, and achieves impressive performance on class imbalance and corrupted label problems where only a small amount of clean validation data is available.

cs.LG↗

arXiv2019

MedGAN: Medical Image Translation using GANs

Image-to-image translation is considered a new frontier in the field of medical image analysis, with numerous potential applications. However, a large portion of recent approaches offers individualized solutions based on specialized task-specific architectures or require refinement through non-end-to-end training. In this paper, we propose a new framework, named MedGAN, for medical image-to-image translation which operates on the image level in an end-to-end manner. MedGAN builds upon recent advances in the field of generative adversarial networks (GANs) by merging the adversarial framework with a new combination of non-adversarial losses. We utilize a discriminator network as a trainable feature extractor which penalizes the discrepancy between the translated medical images and the desired modalities. Moreover, style-transfer losses are utilized to match the textures and fine-structures of the desired target images to the translated images. Additionally, we present a new generator architecture, titled CasNet, which enhances the sharpness of the translated medical outputs through progressive refinement via encoder-decoder pairs. Without any application-specific modifications, we apply MedGAN on three different tasks: PET-CT translation, correction of MR motion artefacts and PET image denoising. Perceptual analysis by radiologists and quantitative evaluations illustrate that the MedGAN outperforms other existing translation approaches.

cs.CV↗

arXiv2019

Temperature dependence of normalized sensitivity of Love wave sensor with unidirectional carbon fiber epoxy composite/Mn-doped 0.24PIN-0.46PMN-0.30PT ternary single crystal configuration

We have derived a general formula for sensitivity optimization of gravimetric sensors and use it to design a high precision and high sensitivity gravimetric sensor using unidirectional carbon fiber epoxy composite (CFEC) guiding layer on single crystal Mn-doped yPb(In1/2Nb1/2)O3-(1-x-y)Pb(Mg1/3Nb2/3)O3-xPbTiO3 (Mn: PIN-PMN-PT) piezoelectric substrate. The normalized maximum sensitivity exhibits a decreasing tendency with temperature up to 55 degrees Celsius. For the CFEC-on-Mn: PIN-PMN-PT sensor configuration with wavelength 24 {mu}m at 25 degrees Celsius, the maximum sensitivity can reach as high as 760.88 cm2/g, which is nearly twice that of traditional SiO2/ST quartz configuration gravimetric sensor.

physics.app-ph↗

arXiv2019

High Resolution Thermal Infrared Imaging of 3200 Phaethon

We present thermal infrared observations of the active asteroid (and Geminid meteoroid stream parent) 3200 Phaethon using the Very Large Telescope. The images, at 10.7 micron wavelength, were taken with Phaethon at its closest approach to Earth (separation 0.07 AU) in 2017 December, at a linear resolution of about 14 km. We probe the Hill sphere (of radius 66 km) for trapped dust and macroscopic bodies, finding neither, and we set limits to the presence of unbound dust. The derived limits to the optical depth of dust near Phaethon depend somewhat on the assumed geometry, but are of order 1e-5. The upper limit to the rate of loss of mass in dust is 14 kg/s. This is about 50 times smaller than the rate needed to sustain the Geminid meteoroid stream in steady state. The observations thus show that the production of the Geminids does not proceed in steady state.

astro-ph.EP↗

arXiv2019

Unsupervised Medical Image Translation Using Cycle-MedGAN

Image-to-image translation is a new field in computer vision with multiple potential applications in the medical domain. However, for supervised image translation frameworks, co-registered datasets, paired in a pixel-wise sense, are required. This is often difficult to acquire in realistic medical scenarios. On the other hand, unsupervised translation frameworks often result in blurred translated images with unrealistic details. In this work, we propose a new unsupervised translation framework which is titled Cycle-MedGAN. The proposed framework utilizes new non-adversarial cycle losses which direct the framework to minimize the textural and perceptual discrepancies in the translated images. Qualitative and quantitative comparisons against other unsupervised translation approaches demonstrate the performance of the proposed framework for PET-CT translation and MR motion correction.

cs.CV↗

arXiv2019

PIXOR: Real-time 3D Object Detection from Point Clouds

We address the problem of real-time 3D object detection from point clouds in the context of autonomous driving. Computation speed is critical as detection is a necessary component for safety. Existing approaches are, however, expensive in computation due to high dimensionality of point clouds. We utilize the 3D data more efficiently by representing the scene from the Bird's Eye View (BEV), and propose PIXOR, a proposal-free, single-stage detector that outputs oriented 3D object estimates decoded from pixel-wise neural network predictions. The input representation, network architecture, and model optimization are especially designed to balance high accuracy and real-time efficiency. We validate PIXOR on two datasets: the KITTI BEV object detection benchmark, and a large-scale 3D vehicle detection benchmark. In both datasets we show that the proposed detector surpasses other state-of-the-art methods notably in terms of Average Precision (AP), while still runs at >28 FPS.

cs.CV↗

arXiv2019

Person Identification and Body Mass Index: A Deep Learning-Based Study on Micro-Dopplers

Obtaining a smart surveillance requires a sensing system that can capture accurate and detailed information for the human walking style. The radar micro-Doppler ($\boldsymbolμ$-D) analysis is proved to be a reliable metric for studying human locomotions. Thus, $\boldsymbolμ$-D signatures can be used to identify humans based on their walking styles. Additionally, the signatures contain information about the radar cross section (RCS) of the moving subject. This paper investigates the effect of human body characteristics on human identification based on their $\boldsymbolμ$-D signatures. In our proposed experimental setup, a treadmill is used to collect $\boldsymbolμ$-D signatures of 22 subjects with different genders and body characteristics. Convolutional autoencoders (CAE) are then used to extract the latent space representation from the $\boldsymbolμ$-D signatures. It is then interpreted in two dimensions using t-distributed stochastic neighbor embedding (t-SNE). Our study shows that the body mass index (BMI) has a correlation with the $\boldsymbolμ$-D signature of the walking subject. A 50-layer deep residual network is then trained to identify the walking subject based on the $\boldsymbolμ$-D signature. We achieve an accuracy of 98% on the test set with high signal-to-noise-ratio (SNR) and 84% in case of different SNR levels.

cs.CV↗

arXiv2019

Hadronic Molecular States Composed of Spin-$3\over 2$ Singly Charmed Baryons

We investigate the possible deuteron-like molecules composed of a pair of charmed spin-$\frac{3}{2}$ baryons, or one charmed baryon and one charmed antibaryon within the one-boson-exchange (OBE) model. For the spin singlet and triplet systems, we consider the couple channel effect between systems with different orbital angular momentum. Most of the systems have binding solutions. The couple channel effect plays a significant role in the formation of some loosely bound states. The possible molecular states of $Ω_c^*Ω_c^*$ and $Ω_c^*\barΩ_c^*$ might be stable once produced.

hep-ph↗