arXiv subjects

Yang Zhao

Publications and source records attributed to Yang Zhao.

At least 397 records · Page 22Linked to original sources

arXiv2020

Where Does It Exist: Spatio-Temporal Video Grounding for Multi-Form Sentences

In this paper, we consider a novel task, Spatio-Temporal Video Grounding for Multi-Form Sentences (STVG). Given an untrimmed video and a declarative/interrogative sentence depicting an object, STVG aims to localize the spatio-temporal tube of the queried object. STVG has two challenging settings: (1) We need to localize spatio-temporal object tubes from untrimmed videos, where the object may only exist in a very small segment of the video; (2) We deal with multi-form sentences, including the declarative sentences with explicit objects and interrogative sentences with unknown objects. Existing methods cannot tackle the STVG task due to the ineffective tube pre-generation and the lack of object relationship modeling. Thus, we then propose a novel Spatio-Temporal Graph Reasoning Network (STGRN) for this task. First, we build a spatio-temporal region graph to capture the region relationships with temporal object dynamics, which involves the implicit and explicit spatial subgraphs in each frame and the temporal dynamic subgraph across frames. We then incorporate textual clues into the graph and develop the multi-step cross-modal graph reasoning. Next, we introduce a spatio-temporal localizer with a dynamic selection method to directly retrieve the spatio-temporal tubes without tube pre-generation. Moreover, we contribute a large-scale video grounding dataset VidSTG based on video relation dataset VidOR. The extensive experiments demonstrate the effectiveness of our method.

cs.CV↗

arXiv2020

Dual-discriminator GAN: A GAN way of profile face recognition

A wealth of angle problems occur when facial recognition is performed: At present, the feature extraction network presents eigenvectors with large differences between the frontal face and profile face recognition of the same person in many cases. For this reason, the state-of-the-art facial recognition network will use multiple samples for the same target to ensure that eigenvector differences caused by angles are ignored during training. However, there is another solution available, which is to generate frontal face images with profile face images before recognition. In this paper, we proposed a method of generating frontal faces with image-to-image profile faces based on Generative Adversarial Network (GAN).

cs.CV↗

arXiv2020

Deep High-Resolution Representation Learning for Visual Recognition

High-resolution representations are essential for position-sensitive vision problems, such as human pose estimation, semantic segmentation, and object detection. Existing state-of-the-art frameworks first encode the input image as a low-resolution representation through a subnetwork that is formed by connecting high-to-low resolution convolutions \emph{in series} (e.g., ResNet, VGGNet), and then recover the high-resolution representation from the encoded low-resolution representation. Instead, our proposed network, named as High-Resolution Network (HRNet), maintains high-resolution representations through the whole process. There are two key characteristics: (i) Connect the high-to-low resolution convolution streams \emph{in parallel}; (ii) Repeatedly exchange the information across resolutions. The benefit is that the resulting representation is semantically richer and spatially more precise. We show the superiority of the proposed HRNet in a wide range of applications, including human pose estimation, semantic segmentation, and object detection, suggesting that the HRNet is a stronger backbone for computer vision problems. All the codes are available at~{\url{https://github.com/HRNet}}.

cs.CV↗

arXiv2020

The Algebraic Expressions of Huygens Principle and Holographic Principle of Light

Huygens principle (HP) is the cornerstone of wave optics, its mathematical model is a boundary value problem of wave equation. The solutions of this mathematical model should be partial derivative u sub n independent and satisfy the form of retarded potential. In the engaged formulas, only the Rayleigh-Sommerfeld diffraction formula (RSDF) satisfies these two restrictions. Unfortunately, the HP requires spherical boundary, while the boundary of RSDF is an infinite plane. Besides that, we find the the geometric constructions of HP and holographic principle of light (HPL) are complementary. Here we derive out the complete expressions of HP and HPL with spherical boundary, based on the method of images. Furthermore, the HP, HPL and RSDF are combined into one new principle that if the boundary of a vacuum region is a spherical surface or an infinite plane, all the light in this vacuum region is determined by the light on the boundary.

physics.class-ph↗

arXiv2019

Learning Diverse Stochastic Human-Action Generators by Learning Smooth Latent Transitions

Human-motion generation is a long-standing challenging task due to the requirement of accurately modeling complex and diverse dynamic patterns. Most existing methods adopt sequence models such as RNN to directly model transitions in the original action space. Due to high dimensionality and potential noise, such modeling of action transitions is particularly challenging. In this paper, we focus on skeleton-based action generation and propose to model smooth and diverse transitions on a latent space of action sequences with much lower dimensionality. Conditioned on a latent sequence, actions are generated by a frame-wise decoder shared by all latent action-poses. Specifically, an implicit RNN is defined to model smooth latent sequences, whose randomness (diversity) is controlled by noise from the input. Different from standard action-prediction methods, our model can generate action sequences from pure noise without any conditional action poses. Remarkably, it can also generate unseen actions from mixed classes during training. Our model is learned with a bi-directional generative-adversarial-net framework, which not only can generate diverse action sequences of a particular class or mix classes, but also learns to classify action sequences within the same model. Experimental results show the superiority of our method in both diverse action-sequence generation and classification, relative to existing methods.

cs.CV↗

arXiv2019

E2-Train: Training State-of-the-art CNNs with Over 80% Energy Savings

Convolutional neural networks (CNNs) have been increasingly deployed to edge devices. Hence, many efforts have been made towards efficient CNN inference in resource-constrained platforms. This paper attempts to explore an orthogonal direction: how to conduct more energy-efficient training of CNNs, so as to enable on-device training. We strive to reduce the energy cost during training, by dropping unnecessary computations from three complementary levels: stochastic mini-batch dropping on the data level; selective layer update on the model level; and sign prediction for low-cost, low-precision back-propagation, on the algorithm level. Extensive simulations and ablation studies, with real energy measurements from an FPGA board, confirm the superiority of our proposed strategies and demonstrate remarkable energy savings for training. For example, when training ResNet-74 on CIFAR-10, we achieve aggressive energy savings of >90% and >60%, while incurring a top-1 accuracy loss of only about 2% and 1.2%, respectively. When training ResNet-110 on CIFAR-100, an over 84% training energy saving is achieved without degrading inference accuracy.

cs.LG↗

arXiv2019

Patchy Image Structure Classification Using Multi-Orientation Region Transform

Exterior contour and interior structure are both vital features for classifying objects. However, most of the existing methods consider exterior contour feature and internal structure feature separately, and thus fail to function when classifying patchy image structures that have similar contours and flexible structures. To address above limitations, this paper proposes a novel Multi-Orientation Region Transform (MORT), which can effectively characterize both contour and structure features simultaneously, for patchy image structure classification. MORT is performed over multiple orientation regions at multiple scales to effectively integrate patchy features, and thus enables a better description of the shape in a coarse-to-fine manner. Moreover, the proposed MORT can be extended to combine with the deep convolutional neural network techniques, for further enhancement of classification accuracy. Very encouraging experimental results on the challenging ultra-fine-grained cultivar recognition task, insect wing recognition task, and large variation butterfly recognition task are obtained, which demonstrate the effectiveness and superiority of the proposed MORT over the state-of-the-art methods in classifying patchy image structures. Our code and three patchy image structure datasets are available at: https://github.com/XiaohanYu-GU/MReT2019.

cs.CV↗

arXiv2019

MobileFAN: Transferring Deep Hidden Representation for Face Alignment

Facial landmark detection is a crucial prerequisite for many face analysis applications. Deep learning-based methods currently dominate the approach of addressing the facial landmark detection. However, such works generally introduce a large number of parameters, resulting in high memory cost. In this paper, we aim for a lightweight as well as effective solution to facial landmark detection. To this end, we propose an effective lightweight model, namely Mobile Face Alignment Network (MobileFAN), using a simple backbone MobileNetV2 as the encoder and three deconvolutional layers as the decoder. The proposed MobileFAN, with only 8% of the model size and lower computational cost, achieves superior or equivalent performance compared with state-of-the-art models. Moreover, by transferring the geometric structural information of a face graph from a large complex model to our proposed MobileFAN through feature-aligned distillation and feature-similarity distillation, the performance of MobileFAN is further improved in effectiveness and efficiency for face alignment. Extensive experiment results on three challenging facial landmark estimation benchmarks including COFW, 300W and WFLW show the superiority of our proposed MobileFAN against state-of-the-art methods.

cs.CV↗

arXiv2019

A Survey of Blockchain Applications in Different Domains

Blockchains have received much attention recently since they provide decentralized approaches to the creation and management of value. Many banks, Internet companies, car manufacturers, and even governments worldwide have incorporated or started considering blockchains to improve the security, scalability, and efficiency of their services. In this paper, we survey blockchain applications in different areas. These areas include cryptocurrency, healthcare, advertising, insurance, copyright protection, energy, and societal applications. Our work provides a timely summary for individuals and organizations interested in blockchains. We envision our study to motivate more blockchain applications.

cs.CR↗

arXiv2019

A Deep Gradient Boosting Network for Optic Disc and Cup Segmentation

Segmentation of optic disc (OD) and optic cup (OC) is critical in automated fundus image analysis system. Existing state-of-the-arts focus on designing deep neural networks with one or multiple dense prediction branches. Such kind of designs ignore connections among prediction branches and their learning capacity is limited. To build connections among prediction branches, this paper introduces gradient boosting framework to deep classification model and proposes a gradient boosting network called BoostNet. Specifically, deformable side-output unit and aggregation unit with deep supervisions are proposed to learn base functions and expansion coefficients in gradient boosting framework. By stacking aggregation units in a deep-to-shallow manner, models' performances are gradually boosted along deep to shallow stages. BoostNet achieves superior results to existing deep OD and OC segmentation networks on the public dataset ORIGA.

cs.CV↗

arXiv2019

Signatures of many-body localization and metastability by weak perturbation

Nonequilibrium dynamics in isolated quantum many-body systems displays a number of intriguing features, such as many-body localization (MBL) and prethermalization. Here we investigate a simple ladder system with disorder, in which various distinct dynamical features coexist and interplay. By exact diagonalization, we demonstrate that the system exhibits the signatures of an MBL-ergodic-MBL reentrant transition, metastability, and disorder-free MBL. We give an account of these properties by introducing a quasi-particle picture and interpreting the quasi-vacuum energy fluctuation as an effective disorder on the quasi-particle dynamics. It is speculated that the weak perturbation behavior is a finite-size effect, but its relaxation time scale increases with the system size.

quant-ph↗

arXiv2019

Variational study of two-impurity spin-boson model with a common Ohmic bath: Ground-state phase transitions

By means of a trial wave function, the multi-D$_1$ ansatz, extensive variational calculations with more than ten thousand parameters have been carried out to study quantum phase transitions in the ground states of a two-impurity system embedded in a common Ohmic bath of bosons. Quantum criticality in both the impurity system and the Ohmic bosonic bath is investigated with relevant transition points and critical exponents determined accurately. With the linear grid of the Ohmic spectral density, our numerical calculations produce a much better description of the ground states with lower energies than other calculations employing a logarithmic grid with a discretization factor far greater than unity. It offers a possible solution to the considerable controversy on the critical coupling in the literature. Moreover, the ground-state phase transition is inferred to be of first order in the presence of strong antiferromagnetic spin-spin couplin}, at variance with that in the ferromagnetic regime or in the absence of spin-spin coupling where the transition belongs to the Kosterlitz-Thouless universality class.

cond-mat.stat-mech↗

arXiv2019

Fast, accurate simulation of polaron dynamics and multidimensional spectroscopy by multiple Davydov trial states

By employing the Dirac-Frenkel time-dependent variational principle, we study the dynamical properties of the Holstein molecular crystal model with diagonal and off-diagonal exciton-phonon coupling. A linear combination of the Davydov D$_1$ (D$_2$) Anstaz, referred to as the multi-D$_1$ Ansatz (multi-D$_2$ Ansatz), is used as the trial state with enhanced accuracy but without sacrificing efficiency. The time evolution of the exciton probability is found to be in perfect agreement with that of the hierarchy equations of motion, demonstrating the promise the multiple Davydov trial states hold as an efficient, robust description of dynamics of complex quantum systems. In addition to the linear absorption spectra computed for both diagonal and off-diagonal cases, for the first time, $2$D spectra have been calculated for systems with off-diagonal exciton-phonon coupling by employing the multiple $D_2$ Ansatz to compute the nonlinear response function, testifying to the great potential of the multiple $D_2$ Ansatz for fast, accurate implementation of multidimensional spectroscopy. It is found that the signal exhibits a single peak for weak off-diagonal coupling, while a vibronic multi-peak structure appears for strong off-diagonal coupling.

cond-mat.stat-mech↗

arXiv2019

An Analysis of Factors Influencing Metro Station Ridership: Insights from Taipei Metro

Travel demand analysis at the planning stage is important for metro system development. In practice, travel demand can be affected by various factors. This paper focuses on investigating the factors influencing Taipei metro ridership at station level over varying time periods. Ordinary Least Square (OLS) multiple regression models with backward stepwise feature selection are employed to identify the influencing factors, including land use, social economic, accessibility, network structure information, etc. Network structure factors are creatively quantified based on complex network theory to accurately measure the related information. To enhance goodness-of-fit, the dummy variable distinguishing transportation hub is incorporated in the modeling. The main findings in this paper are three-fold: First, there is no distinct difference between influencing factors of boarding and those of alighting; Second, ridership is significantly associated with the number of nearby shopping malls, distance to city center, days since opening, nearby bus stations and dummy variable for transportation hub; Finally, the ridership on weekdays is mainly affected by commuting activities, while the ridership on weekends is driven by commercial access.

stat.AP↗

arXiv2019

Dissipative dynamics in a tunable Rabi dimer with periodic harmonic driving

Recent progress on qubit manipulation allows application of periodic driving signals on qubits. In this study, a harmonic driving field is added to a Rabi dimer to engineer photon and qubit dynamics in a circuit quantum electrodynamics device. To model environmental effects, qubits in the Rabi dimer are coupled to a phonon bath with a sub-Ohmic spectral density. A non-perturbative treatment, the Dirac-Frenkel time-dependent variational principle together with the multiple Davydov D$_2$ {\it Ansatz} is employed to explore the dynamical behavior of the tunable Rabi dimer. In the absence of the phonon bath, the amplitude damping of the photon number oscillation is greatly suppressed by the driving field, and photons can be created thanks to resonances between the periodic driving field and the photon frequency. In the presence of the phonon bath, one still can change the photon numbers in two resonators, and indirectly alter the photon imbalance in the Rabi dimer by directly varying the driving signal in one qubit. It is shown that qubit states can be manipulated directly by the harmonic driving. The environment is found to strengthen the interqubit asymmetry induced by the external driving, opening up a new venue to engineer the qubit states.

quant-ph↗

arXiv2019

High-Resolution Representations for Labeling Pixels and Regions

High-resolution representation learning plays an essential role in many vision problems, e.g., pose estimation and semantic segmentation. The high-resolution network (HRNet)~\cite{SunXLW19}, recently developed for human pose estimation, maintains high-resolution representations through the whole process by connecting high-to-low resolution convolutions in \emph{parallel} and produces strong high-resolution representations by repeatedly conducting fusions across parallel convolutions. In this paper, we conduct a further study on high-resolution representations by introducing a simple yet effective modification and apply it to a wide range of vision tasks. We augment the high-resolution representation by aggregating the (upsampled) representations from all the parallel convolutions rather than only the representation from the high-resolution convolution as done in~\cite{SunXLW19}. This simple modification leads to stronger representations, evidenced by superior results. We show top results in semantic segmentation on Cityscapes, LIP, and PASCAL Context, and facial landmark detection on AFLW, COFW, $300$W, and WFLW. In addition, we build a multi-level representation from the high-resolution representation and apply it to the Faster R-CNN object detection framework and the extended frameworks. The proposed approach achieves superior results to existing single-model networks on COCO object detection. The code and models have been publicly available at \url{https://github.com/HRNet}.

cs.CV↗

arXiv2019

An Adapted Geographically Weighted Lasso(Ada-GWL) model for estimating metro ridership

Ridership estimation at station level plays a critical role in metro transportation planning. Among various existing ridership estimation methods, direct demand model has been recognized as an effective approach. However, existing direct demand models including Geographically Weighted Regression (GWR) have rarely included local model selection in ridership estimation. In practice, acquiring insights into metro ridership under multiple influencing factors from a local perspective is important for passenger volume management and transportation planning operations adapting to local conditions. In this study, we propose an Adapted Geographically Weighted Lasso (Ada-GWL) framework for modelling metro ridership, which performs regression-coefficient shrinkage and local model selection. It takes metro network connection intermedia into account and adopts network-based distance metric instead of Euclidean-based distance metric, making it so-called adapted to the context of metro networks. The real-world case of Shenzhen Metro is used to validate the superiority of our proposed model. The results show that the Ada-GWL model performs the best compared with the global model (Ordinary Least Square (OLS), GWR, GWR calibrated with network-based distance metric and GWL in terms of estimation error of the dependent variable and goodness-of-fit. Through understanding the variation of each coefficient across space (elasticities) and variables selection of each station, it provides more realistic conclusions based on local analysis. Besides, through clustering analysis of the stations according to the regression coefficients, clusters' functional characteristics are found to be in compliance with the facts of the functional land use policy of Shenzhen. These results of the proposed Ada-GWL model demonstrate a great spatial explanatory power in transportation planning.

stat.AP↗

arXiv2019

Random Occlusion-recovery for Person Re-identification

As a basic task of multi-camera surveillance system, person re-identification aims to re-identify a query pedestrian observed from non-overlapping multiple cameras or across different time with a single camera. Recently, deep learning-based person re-identification models have achieved great success in many benchmarks. However, these supervised models require a large amount of labeled image data, and the process of manual labeling spends much manpower and time. In this study, we introduce a method to automatically synthesize labeled person images and adopt them to increase the sample number per identity for person re-identification datasets. To be specific, we use block rectangles to randomly occlude pedestrian images. Then, a generative adversarial network (GAN) model is proposed to use paired occluded and original images to synthesize the de-occluded images that similar but not identical to the original image. Afterwards, we annotate the de-occluded images with the same labels of their corresponding raw images and use them to augment the number of samples per identity. Finally, we use the augmented datasets to train baseline model. The experiment results on CUHK03, Market-1501 and DukeMTMC-reID datasets show that the effectiveness of the proposed method.

cs.CV↗