arXiv subjects

Ying Wu

Publications and source records attributed to Ying Wu.

At least 91 records · Page 5Linked to original sources

arXiv2021

Morphable Detector for Object Detection on Demand

Many emerging applications of intelligent robots need to explore and understand new environments, where it is desirable to detect objects of novel classes on the fly with minimum online efforts. This is an object detection on demand (ODOD) task. It is challenging, because it is impossible to annotate a large number of data on the fly, and the embedded systems are usually unable to perform back-propagation which is essential for training. Most existing few-shot detection methods are confronted here as they need extra training. We propose a novel morphable detector (MD), that simply "morphs" some of its changeable parameters online estimated from the few samples, so as to detect novel classes without any extra training. The MD has two sets of parameters, one for the feature embedding and the other for class representation (called "prototypes"). Each class is associated with a hidden prototype to be learned by integrating the visual and semantic embeddings. The learning of the MD is based on the alternate learning of the feature embedding and the prototypes in an EM-like approach which allows the recovery of an unknown prototype from a few samples of a novel class. Once an MD is learned, it is able to use a few samples of a novel class to directly compute its prototype to fulfill the online morphing process. We have shown the superiority of the MD in Pascal, COCO and FSOD datasets. The code is available https://github.com/Zhaoxiangyun/Morphable-Detector.

cs.CV↗

arXiv2021

Contrastive Learning for Label-Efficient Semantic Segmentation

Collecting labeled data for the task of semantic segmentation is expensive and time-consuming, as it requires dense pixel-level annotations. While recent Convolutional Neural Network (CNN) based semantic segmentation approaches have achieved impressive results by using large amounts of labeled training data, their performance drops significantly as the amount of labeled data decreases. This happens because deep CNNs trained with the de facto cross-entropy loss can easily overfit to small amounts of labeled data. To address this issue, we propose a simple and effective contrastive learning-based training strategy in which we first pretrain the network using a pixel-wise, label-based contrastive loss, and then fine-tune it using the cross-entropy loss. This approach increases intra-class compactness and inter-class separability, thereby resulting in a better pixel classifier. We demonstrate the effectiveness of the proposed training strategy using the Cityscapes and PASCAL VOC 2012 segmentation datasets. Our results show that pretraining with the proposed contrastive loss results in large performance gains (more than 20% absolute improvement in some settings) when the amount of labeled data is limited. In many settings, the proposed contrastive pretraining strategy, which does not use any additional data, is able to match or outperform the widely-used ImageNet pretraining strategy that uses more than a million additional labeled images.

cs.CV↗

arXiv2021

Parity-Symmetry-Protected Bundle Emission

We demonstrate symmetry protected bundle emission in the cavity QED system under the ultrastrong coupling regime. Parity symmetry of Rabi model only permits the super-Rabi oscillations with periodic generation of even photons in cavity, which is realized by the laser driven flip of qubit and the parity conserved transitions induced by Rabi interaction. Combined with dissipation, only 2$n$-photon correlated emissions are allowed, and the corresponding purities are significantly enhanced by the parity symmetry, due to the almost perfect suppression of bundle emissions with odd correlated photons. The switching of bundle emissions between even and odd correlated photons can be controlled by manipulating the parity symmetry of system with an added magnetic field. This work extends multi-photon bundle emission to the ultrastrong coupling regime, and offers the prospect of exploring symmetry-protected multi-quanta physics.

quant-ph↗

arXiv2021

Experimental realization of single-plaquette gauge flux insertion and topological Wannier cycles

Gauge fields are at the heart of the fundamental science of our universe and various materials. For instance, Laughlin's gedanken experiment of gauge flux insertion played a major role in understanding the quantum Hall effects. Gauge flux insertion into a single unit-cell, though crucial for detecting exotic quantum phases and for the ultimate control of quantum dynamics and classical waves, however, has not yet been achieved in laboratory. Here, we report on the experimental realization of gauge flux insertion into a single plaquette in a lattice system with the gauge phase ranging from 0 to 2pi which is realized through a novel approach based on three consecutive procedures: the dimension extension, creating an engineered dislocation and the dimensional reduction. Furthermore, we discover that the single-plaquette gauge flux insertion leads to a new phenomenon termed as the topological Wannier cycles, i.e., the cyclic spectral flows across multiple band gaps which are manifested as the topological boundary states (TBSs) on the plaquette. Such topological Wannier cycles emerge only if the Wannier centers are enclosed by the flux-carrying plaquette. Exploiting acoustic metamaterials and versatile pump-probe measurements, we observe the topological Wannier cycles by detecting the TBSs in various ways and confirm the single-plaquette gauge flux insertion by measuring the gauge phase accumulation on the plaquette. Our work unveils an unprecedented regime for lattice gauge systems and a fundamental topological response which could empower future studies on artificial gauge fields and topological materials.

cond-mat.mtrl-sci↗

arXiv2021

Unsupervised Embedding Learning from Uncertainty Momentum Modeling

Existing popular unsupervised embedding learning methods focus on enhancing the instance-level local discrimination of the given unlabeled images by exploring various negative data. However, the existed sample outliers which exhibit large intra-class divergences or small inter-class variations severely limit their learning performance. We justify that the performance limitation is caused by the gradient vanishing on these sample outliers. Moreover, the shortage of positive data and disregard for global discrimination consideration also pose critical issues for unsupervised learning but are always ignored by existing methods. To handle these issues, we propose a novel solution to explicitly model and directly explore the uncertainty of the given unlabeled learning samples. Instead of learning a deterministic feature point for each sample in the embedding space, we propose to represent a sample by a stochastic Gaussian with the mean vector depicting its space localization and covariance vector representing the sample uncertainty. We leverage such uncertainty modeling as momentum to the learning which is helpful to tackle the outliers. Furthermore, abundant positive candidates can be readily drawn from the learned instance-specific distributions which are further adopted to mitigate the aforementioned issues. Thorough rationale analyses and extensive experiments are presented to verify our superiority.

cs.CV↗

arXiv2021

Beyond Visual Attractiveness: Physically Plausible Single Image HDR Reconstruction for Spherical Panoramas

HDR reconstruction is an important task in computer vision with many industrial needs. The traditional approaches merge multiple exposure shots to generate HDRs that correspond to the physical quantity of illuminance of the scene. However, the tedious capturing process makes such multi-shot approaches inconvenient in practice. In contrast, recent single-shot methods predict a visually appealing HDR from a single LDR image through deep learning. But it is not clear whether the previously mentioned physical properties would still hold, without training the network to explicitly model them. In this paper, we introduce the physical illuminance constraints to our single-shot HDR reconstruction framework, with a focus on spherical panoramas. By the proposed physical regularization, our method can generate HDRs which are not only visually appealing but also physically plausible. For evaluation, we collect a large dataset of LDR and HDR images with ground truth illuminance measures. Extensive experiments show that our HDR images not only maintain high visual quality but also top all baseline methods in illuminance prediction accuracy.

cs.CV↗

arXiv2021

MetaView: Few-shot Active Object Recognition

In robot sensing scenarios, instead of passively utilizing human captured views, an agent should be able to actively choose informative viewpoints of a 3D object as discriminative evidence to boost the recognition accuracy. This task is referred to as active object recognition. Recent works on this task rely on a massive amount of training examples to train an optimal view selection policy. But in realistic robot sensing scenarios, the large-scale training data may not exist and whether the intelligent view selection policy can be still learned from few object samples remains unclear. In this paper, we study this new problem which is extremely challenging but very meaningful in robot sensing -- Few-shot Active Object Recognition, i.e., to learn view selection policies from few object samples, which has not been considered and addressed before. We solve the proposed problem by adopting the framework of meta learning and name our method "MetaView". Extensive experiments on both category-level and instance-level classification tasks demonstrate that the proposed method can efficiently resolve issues that are hard for state-of-the-art active object recognition methods to handle, and outperform several baselines by large margins.

cs.RO↗

arXiv2021

Segregation, integration and balance of large-scale resting brain networks configure different cognitive abilities

Diverse cognitive processes set different demands on locally segregated and globally integrated brain activity. However, it remains unclear how resting brains configure their functional organization to balance the demands on network segregation and integration to best serve cognition. Here, we use an eigenmode-based approach to identify hierarchical modules in functional brain networks, and quantify the functional balance between network segregation and integration. In a large sample of healthy young adults (n=991), we combine the whole-brain resting state functional magnetic resonance imaging (fMRI) data with a mean-filed model on the structural network derived from diffusion tensor imaging and demonstrate that resting brain networks are on average close to a balanced state. This state allows for a balanced time dwelling at segregated and integrated configurations, and highly flexible switching between them. Furthermore, we employ structural equation modelling to estimate general and domain-specific cognitive phenotypes from nine tasks, and demonstrate that network segregation, integration and their balance in resting brains predict individual differences in diverse cognitive phenotypes. More specifically, stronger integration is associated with better general cognitive ability, stronger segregation fosters crystallized intelligence and processing speed, and individual's tendency towards balance supports better memory. Our findings provide a comprehensive and deep understanding of the brain's functioning principles in supporting diverse functional demands and cognitive abilities, and advance modern network neuroscience theories of human cognition.

q-bio.NC↗

arXiv2021

Multi-dimensional wave steering with higher-order topological phononic crystal

The recent discovery and realizations of higher-order topological insulators enrich the fundamental studies on topological phases. Here, we report three-dimensional (3D) wave-steering capabilities enabled by topological boundary states at three different orders in a 3D phononic crystal with nontrivial bulk topology originated from the synergy of mirror symmetry of the unit cell and a non-symmorphic glide symmetry of the lattice. The multitude of topological states brings diverse possibility of wave manipulations. Through judicious engineering of the boundary modes, we experimentally demonstrate two functionalities at different dimensions: 2D negative refraction of sound wave enabled by a first-order topological surface state with negative dispersion, and a 3D acoustic interferometer leveraging on second-order topological hinge states. Our work showcases that topological modes at different orders promise diverse wave steering applications across different dimensions.

physics.app-ph↗

arXiv2021

Abnormal topological refraction into free medium at sub-wavelength scale in valley phononic crystal plates

In this work we propose a topological valley phononic crystal plate and we extensively investigate the refraction of valley modes into the surrounding homogeneous medium. This phononic crystal includes two sublattices of resonators (A and B) modeled by mass-spring systems. We show that two edge states confined at the AB/BA and BA/AB type domain walls exhibit different symmetries in physical space and energy peaks in the Fourier space. As a result, distinct refraction behaviors, especially through an armchair cut edge, are observed. On the other hand, the decay depth of these localized topological modes, which is found to be solely determined by the relative resonant strength between the scatterers, significantly affects the refraction patterns. More interestingly, the outgoing traveling wave through a zigzag interface becomes evanescent when operating at deep sub-wavelength scale. This is realized by tuning the average resonant strength. We show that the evanescent modes only exist along a particular type of outlet edge, and that they can couple with both topological interface states. We also present two designs of topological functional devices, including an elastic one-way transmission waveguide and a near-ideal monopole/dipole emitter, both based on our phononic structure.

physics.app-ph↗

arXiv2020

Three-dimensional Electromagnetic Void Space

We report a realization of three-dimensional (3D) electromagnetic void space. Despite occupying a finite volume of space, such a medium is optically equivalent to an infinitesimal point where electromagnetic waves experience no phase accumulation. The 3D void space is realized by constructing all-dielectric 3D photonic crystals such that the effective permittivity and permeability vanish simultaneously, forming a six-fold Dirac-like point with Dirac-like linear dispersions at the center of the Brillouin Zone. We demonstrate, both theoretically and experimentally, that such a 3D void space exhibits unique properties and rich functionalities absent in any other electromagnetic media, such as boundary-control transmission switching and 3D perfect wave-steering mechanisms. Especially, contrary to the photonic "doping" effect in its two-dimensional counterpart, the 3D void space exhibits an amazing property of "impurity-immunity". Our work paves a road towards the realization of 3D void space where electromagnetic waves can be manipulated in unprecedented ways.

physics.optics↗

arXiv2020

Highly nonclassical phonon emission statistics through two-phonon loss of van der Pol oscillator

The ability to produce nonclassical wave in a system is essential for advances in quantum communication and computation. Here, we propose a scheme to generate highly nonclassical phonon emission statistics -- antibunched wave in a quantum van der Pol (vdP) oscillator subject to an external driving, both single- and two-phonon losses. It is found that phonon antibunching depends significantly on the nonlinear two-phonon loss of the vdP oscillator, where the degree of the antibunching increases monotonically with the two-phonon loss and the distinguished parameter regimes with optimal antibunching and single-phonon emission are identified clearly. In addition, we give an in-depth insight into strong antibunching in the emitted phonon statistics by analytical calculations using a three-oscillator-level model, which agree well with the full numerical simulations employing a master-equation approach and a Schrodinger-equation approach at weak driving. In turn, the fluorescence phonon emission spectra of the vdP oscillator, given by the power spectral density, are also evaluated. We further show that high phonon emission amplitudes, simultaneously accompanied by strong phonon antibunching, are attainable in the vdP system, which are beneficial to the correlation measurement in practical experiments. Our approach only requires a single vdP oscillator, without the need for reconfiguring the two coupled nonlinear resonators or the complex nanophotonic structures as compared to the previous unconventional blockade schemes. The present scheme could inspire methods to achieve antibunching in other systems.

physics.optics↗

arXiv2020

Observation of a phononic higher-order Weyl semimetal

Weyl semimetals are extraordinary systems where exotic phenomena such as Fermi arcs, pseudo-gauge fields and quantum anomalies arise from topological band degeneracy in crystalline solids for electrons and metamaterials for photons and phonons. On the other hand, higher-order topological insulators unveil intriguing multidimensional topological physics beyond the conventional bulk-edge correspondences. However, it is unclear whether higher-order topology can emerge in Weyl semimetals. Here, we report the experimental discovery of higher-order Weyl semimetals in its phononic analog which exhibit topologically-protected boundary states in multiple dimensions. We create the physical realization of the higher-order Weyl semimetal in a chiral phononic crystal with uniaxial screw symmetry. Using near-field spectroscopies, we observe the chiral Fermi arcs on the surfaces and a new type of hinge arc states on the hinge boundaries. These topological boundary arc states link the projections of Weyl points in different dimensions and directions, and hence demonstrate higher-order multidimensional topological physics in Weyl semimetals. Our study establishes the fundamental connection between higher-order topology and Weyl physics in crystalline materials and unveils a new horizon of higher-order topological semimetals where unprecedented materials such as higher-order topological nodal-lines may emerge.

cond-mat.mes-hall↗

arXiv2020

Deterministic and probabilistic deep learning models for inverse design of broadband acoustic cloak

Concealing an object from incoming waves (light and/or sound) remained science fiction for a long time due to the absence of wave-shielding materials in nature. Yet, the invention of artificial materials and new physical principles for optical and sound wave manipulation translated this abstract concept into reality by making an object acoustically invisible. Here, we present the notion of a machine learning-driven acoustic cloak and demonstrate an example of such a cloak with a multilayered core-shell configuration. Importantly, we develop deterministic and probabilistic deep learning models based on autoencoder-like neural network structure to retrieve the structural and material properties of the cloaking shell surrounding the object that suppresses scattering of sound in a broad spectral range, as if it was not there. The probabilistic model enhances the generalization ability of design procedure and uncovers the sensitivity of the cloak parameters on the spectral response for practical implementation. This proposal opens up new avenues to expedite the design of intelligent cloaking devices for tailored spectral response and offers a feasible solution for inverse scattering problems.

physics.app-ph↗

arXiv2020

On-chip higher-order topological micromechanical metamaterials

Higher-order topological insulators exhibit multidimensional topological physics and unique application values due to their ability of integrating stable boundary states at multiple dimensions in a single chip. However, for signal-processing applications in high-frequency mechanical systems, the current realizations of higher-order topological mechanical materials are still limited to large-scale systems for kilohertz or lower frequencies. Here, we report the experimental observation of a on-chip micromechanical metamaterial as higher-order topological insulator for high-frequency mechanical waves. The higher-order topological phononic band gap is induced by the band inversion at the Brillouin zone corner which is achieved by configuring the orientations of the elliptic pillars etched on the silicon chip. With consistent experiments and theory, we demonstrate the coexistence of topological edge and corner states in the megahertz frequency regime as induced by the higher-order band topology. The experimental realization of on-chip micromechanical metamaterials with higher-order topology opens a regime for applications based on megahertz mechanical waves in an integrated platform where the edge and corner states act as stable waveguides and resonators, respectively.

physics.app-ph↗

arXiv2020

Interplay of quantum phase transition and flat band in hybrid lattices

We investigate the interplay of superradiant phase transition (SPT) and energy band physics in an extended Dicke-Hubbard lattice whose unit cell consists of a Dicke model coupled to an atomless cavity. We found in such a periodic lattice the critical point that occurs in a single Dicke model becomes a critical region that is periodically changing with the wavenumber $k$. In the weak-coupling normal phase of the system we observed a flat band and its corresponding localization that can be controlled by the ground-state SPT. Our work builds the connection between flat band physics and SPT, which may fundamentally broaden the regimes of many-body theory and quantum optics.

quant-ph↗

arXiv2020

Object Detection with a Unified Label Space from Multiple Datasets

Given multiple datasets with different label spaces, the goal of this work is to train a single object detector predicting over the union of all the label spaces. The practical benefits of such an object detector are obvious and significant application-relevant categories can be picked and merged form arbitrary existing datasets. However, naive merging of datasets is not possible in this case, due to inconsistent object annotations. Consider an object category like faces that is annotated in one dataset, but is not annotated in another dataset, although the object itself appears in the latter images. Some categories, like face here, would thus be considered foreground in one dataset, but background in another. To address this challenge, we design a framework which works with such partial annotations, and we exploit a pseudo labeling approach that we adapt for our specific case. We propose loss functions that carefully integrate partial but correct annotations with complementary but noisy pseudo labels. Evaluation in the proposed novel setting requires full annotation on the test set. We collect the required annotations and define a new challenging experimental setup for this task based one existing public datasets. We show improved performances compared to competitive baselines and appropriate adaptations of existing work.

cs.CV↗

arXiv2020

Exposing Deep-faked Videos by Anomalous Co-motion Pattern Detection

Recent deep learning based video synthesis approaches, in particular with applications that can forge identities such as "DeepFake", have raised great security concerns. Therefore, corresponding deep forensic methods are proposed to tackle this problem. However, existing methods are either based on unexplainable deep networks which greatly degrades the principal interpretability factor to media forensic, or rely on fragile image statistics such as noise pattern, which in real-world scenarios can be easily deteriorated by data compression. In this paper, we propose an fully-interpretable video forensic method that is designed specifically to expose deep-faked videos. To enhance generalizability on videos with various content, we model the temporal motion of multiple specific spatial locations in the videos to extract a robust and reliable representation, called Co-Motion Pattern. Such kind of conjoint pattern is mined across local motion features which is independent of the video contents so that the instance-wise variation can also be largely alleviated. More importantly, our proposed co-motion pattern possesses both superior interpretability and sufficient robustness against data compression for deep-faked videos. We conduct extensive experiments to empirically demonstrate the superiority and effectiveness of our approach under both classification and anomaly detection evaluation settings against the state-of-the-art deep forensic methods.

cs.CV↗