arXiv subjects

Qin Xu

Publications and source records attributed to Qin Xu.

At least 37 records · Page 2Linked to original sources

arXiv2023

Robust Transductive Few-shot Learning via Joint Message Passing and Prototype-based Soft-label Propagation

Few-shot learning (FSL) aims to develop a learning model with the ability to generalize to new classes using a few support samples. For transductive FSL tasks, prototype learning and label propagation methods are commonly employed. Prototype methods generally first learn the representative prototypes from the support set and then determine the labels of queries based on the metric between query samples and prototypes. Label propagation methods try to propagate the labels of support samples on the constructed graph encoding the relationships between both support and query samples. This paper aims to integrate these two principles together and develop an efficient and robust transductive FSL approach, termed Prototype-based Soft-label Propagation (PSLP). Specifically, we first estimate the soft-label presentation for each query sample by leveraging prototypes. Then, we conduct soft-label propagation on our learned query-support graph. Both steps are conducted progressively to boost their respective performance. Moreover, to learn effective prototypes for soft-label estimation as well as the desirable query-support graph for soft-label propagation, we design a new joint message passing scheme to learn sample presentation and relational graph jointly. Our PSLP method is parameter-free and can be implemented very efficiently. On four popular datasets, our method achieves competitive results on both balanced and imbalanced settings compared to the state-of-the-art methods. The code will be released upon acceptance.

cs.CV↗

arXiv2023

The dissipativity and controllability of input affine systems

The paper is considered with the dissipative theory and feedback control under the framework of dissipation with the supply rate is the inner product of input u and the derivation of output y for the linear/nonlinear time-invariant input affine system. A necessary and sufficient condition of dissipativity of the class of systems is presented. Furthermore, we obtain necessary and sufficient conditions of the negative imaginary property, input strictly negative imaginary property and output strictly negative imaginary property for the nonlinear input affine system, which can cover the case that the linear time-invariant system in the literature. Besides, under given conditions, we consider the equivalence between the negative imaginary property and input-output stability. In other words, the system with negative imaginary property may be input-output stable under some certain conditions; conversely, the input-output stable system may also have negative imaginary property. Last but not least, we consider the equivalence between the asymptotical stability and strict dissipativity with the supply rate and some dissipative matrices are given for a certain linear time-invariant system.

math.DS↗

arXiv2022

Strong photon-magnon coupling using a lithographically defined organic ferrimagnet

We demonstrate a hybrid quantum system composed of superconducting resonator photons and magnons hosted by the organic-based ferrimagnet vanadium tetracyanoethylene (V[TCNE]$_x$). Our work is motivated by the challenge of scalably integrating an arbitrarily-shaped, low-damping magnetic system with planar superconducting circuits, thus enabling a host of quantum magnonic circuit designs that were previously inaccessible. For example, by leveraging the inherent properties of magnons, one can enable nonreciprocal magnon-mediated quantum devices that use magnon propagation rather than electrical current. We take advantage of the properties of V[TCNE]$_x$, which has ultra-low intrinsic damping, can be grown at low processing temperatures on arbitrary substrates, and can be patterned via electron beam lithography. We demonstrate the scalable, lithographically integrated fabrication of hybrid quantum magnonic devices consisting of a thin-film superconducting resonator coupled to a low-damping, thin-film V[TCNE]$_x$ microstructure. Our devices operate in the strong coupling regime, with a cooperativity as high as 1181(44) at T$\sim$0.4 K, suitable for scalable quantum circuit integration. This work paves the way for the exploration of high-cooperativity hybrid magnonic quantum devices in which magnonic circuits can be designed and fabricated as easily as electrical wires.

quant-ph↗

arXiv2021

The role of crosslinking density in surface stress and surface energy of soft solids

Surface stress and surface energy are two fundamental parameters that determine the surface properties of any materials. While it is commonly believed that the surface stress and surface energy of liquids are identical, the relationship between the two parameters in soft polymeric gels remains debatable. In this work, we measured the surface stress and surface energy of soft silicone gels with varying weight ratios of crosslinkers in soft wetting experiments. Above a critical density, $k_0$, the surface stress was found to increase significantly with crosslinking density while the surface energy remained unchanged. In this regime, we can estimate a non-zero surface elastic modulus that also increases with the ratio of crosslinkers. By comparing the surface mechanics of the soft gels with their bulk rheology, the surface properties near the critical density $k_0$ were found to be closely related to the underlying percolation transition of the polymer networks.

cond-mat.soft↗

arXiv2021

Droplets sit and slide anisotropically on soft, stretched substrates

Anisotropically wetting substrates enable useful control of droplet behavior across a range of applications. Usually, these involve chemically or physically patterning the substrate surface, or applying gradients in properties like temperature or electrical field. Here, we show that a flat, stretched, uniform soft substrate also exhibits asymmetric wetting, both in terms of how droplets slide and in their static shape. Droplet dynamics are strongly affected by stretch: glycerol droplets on silicone substrates with a 23\% stretch slide 67\% faster in the direction parallel to the applied stretch than in the perpendicular direction. Contrary to classical wetting theory, static droplets in equilibrium appear elongated, oriented parallel to the stretch direction. Both effects arise from droplet-induced deformations of the substrate near the contact line.

cond-mat.soft↗

arXiv2020

Viscoelastic and poroelastic relaxations of soft solid surfaces

Understanding surface mechanics of soft solids, such as soft polymeric gels, is crucial in many engineering processes, such as dynamic wetting and adhesive failure. In these situations, a combination of capillary and elastic forces drives the motion, which is balanced by dissipative mechanisms to determine the rate. While shear rheology (i.e. viscoelasticity) has long been assumed to dominate the dissipation, recent works have suggested that compressibility effects (i.e. poroelasticity) could play roles in swollen networks. We use fast interferometric imaging to quantify the relaxation of surface deformations due to a displaced contact line. By systematically measuring the profiles at different time and length scales, we experimentally observe a crossover from viscoelastic to poroelastic surface relaxations.

cond-mat.soft↗

arXiv2020

Practical Deep Raw Image Denoising on Mobile Devices

Deep learning-based image denoising approaches have been extensively studied in recent years, prevailing in many public benchmark datasets. However, the stat-of-the-art networks are computationally too expensive to be directly applied on mobile devices. In this work, we propose a light-weight, efficient neural network-based raw image denoiser that runs smoothly on mainstream mobile devices, and produces high quality denoising results. Our key insights are twofold: (1) by measuring and estimating sensor noise level, a smaller network trained on synthetic sensor-specific data can out-perform larger ones trained on general data; (2) the large noise level variation under different ISO settings can be removed by a novel k-Sigma Transform, allowing a small network to efficiently handle a wide range of noise levels. We conduct extensive experiments to demonstrate the efficiency and accuracy of our approach. Our proposed mobile-friendly denoising model runs at ~70 milliseconds per megapixel on Qualcomm Snapdragon 855 chipset, and it is the basis of the night shot feature of several flagship smartphones released in 2019.

eess.IV↗

arXiv2020

Contact lines on stretched soft solids: Modeling anisotropic surface stresses

We present fully analytical solutions for the deformation of a stretched soft substrate due to the static wetting of a large liquid droplet, and compare our solutions to recently published experiments (Xu et al, Soft Matter 2018). Following a Green's function approach, we extend the surface-stress regularized Flamant-Cerruti problem to account for uniaxial pre-strains of the substrate. Surface profiles, including the heights and opening angles of wetting ridges, are provided for linearized and finite kinematics. We fit experimental wetting ridge shapes as a function of applied strain using two free parameters, the surface Lame coefficients. In comparison with experiments, we find that observed opening angles are more accurately captured using finite kinematics, especially with increasing levels of applied pre-strain. These fits qualitatively agree with the results of Xu et al, but revise values of the surface elastic constants.

cond-mat.soft↗

arXiv2020

A Guaranteed Convergence Analysis for the Projected Fast Iterative Soft-Thresholding Algorithm in Parallel MRI

The boom of non-uniform sampling and compressed sensing techniques dramatically alleviates the lengthy data acquisition problem of magnetic resonance imaging. Sparse reconstruction, thanks to its fast computation and promising performance, has attracted researchers to put numerous efforts on it and has been adopted in commercial scanners. To perform sparse reconstruction, choosing a proper algorithm is essential in providing satisfying results and saving time in tuning parameters. The pFISTA, a simple and efficient algorithm for sparse reconstruction, has been successfully extended to parallel imaging. However, its convergence criterion is still an open question. And the existing convergence criterion of single-coil pFISTA cannot be applied to the parallel imaging pFISTA, which, therefore, imposes confusions and difficulties on users about determining the only parameter - step size. In this work, we provide the guaranteed convergence analysis of the parallel imaging version pFISTA to solve the two well-known parallel imaging reconstruction models, SENSE and SPIRiT. Along with the convergence analysis, we provide recommended step size values for SENSE and SPIRiT reconstructions to obtain fast and promising reconstructions. Experiments on in vivo brain images demonstrate the validity of the convergence criterion. Besides, experimental results show that compared to using backtracking and power iteration to determine the step size, our recommended step size achieves more than five times acceleration in reconstruction time in most tested cases.

eess.IV↗

arXiv2019

Stress fluctuations and shear thickening in dense granular suspensions

We experimentally investigate the rheology and stress fluctuations of granules densely suspended in silicone oil. We find that both thickening strength and stress fluctuations significantly weaken with oil viscosity $η_0$. Comparison of our rheological results to the Wyart-Cates model for describing different dynamic jamming states suggests a transition from frictional contacts to lubrication interactions as $η_0$ increases. To clarify the contribution from viscous interactions to the rheology, we systematically measure stress fluctuations in various flow states. Reduction of stress fluctuations with $η_0$ indicates that a strong lubrication layer greatly inhibits force correlations among particles. Measuring stress fluctuations in the strong shear thickening regime, we observe a crossover from asymmetric Gamma to symmetric Gaussian distributions and associated with it a decrease of lateral (radial) correlation length $ξ$ with increasing shear rate.

cond-mat.soft↗

arXiv2019

Learning Raw Image Denoising with Bayer Pattern Unification and Bayer Preserving Augmentation

In this paper, we present new data pre-processing and augmentation techniques for DNN-based raw image denoising. Compared with traditional RGB image denoising, performing this task on direct camera sensor readings presents new challenges such as how to effectively handle various Bayer patterns from different data sources, and subsequently how to perform valid data augmentation with raw images. To address the first problem, we propose a Bayer pattern unification (BayerUnify) method to unify different Bayer patterns. This allows us to fully utilize a heterogeneous dataset to train a single denoising model instead of training one model for each pattern. Furthermore, while it is essential to augment the dataset to improve model generalization and performance, we discovered that it is error-prone to modify raw images by adapting augmentation methods designed for RGB images. Towards this end, we present a Bayer preserving augmentation (BayerAug) method as an effective approach for raw image augmentation. Combining these data processing technqiues with a modified U-Net, our method achieves a PSNR of 52.11 and a SSIM of 0.9969 in NTIRE 2019 Real Image Denoising Challenge, demonstrating the state-of-the-art performance. Our code is available at https://github.com/Jiaming-Liu/BayerUnifyAug.

cs.CV↗

arXiv2019

Perceive Where to Focus: Learning Visibility-aware Part-level Features for Partial Person Re-identification

This paper considers a realistic problem in person re-identification (re-ID) task, i.e., partial re-ID. Under partial re-ID scenario, the images may contain a partial observation of a pedestrian. If we directly compare a partial pedestrian image with a holistic one, the extreme spatial misalignment significantly compromises the discriminative ability of the learned representation. We propose a Visibility-aware Part Model (VPM), which learns to perceive the visibility of regions through self-supervision. The visibility awareness allows VPM to extract region-level features and compare two images with focus on their shared regions (which are visible on both images). VPM gains two-fold benefit toward higher accuracy for partial re-ID. On the one hand, compared with learning a global feature, VPM learns region-level features and benefits from fine-grained information. On the other hand, with visibility awareness, VPM is capable to estimate the shared regions between two images and thus suppresses the spatial misalignment. Experimental results confirm that our method significantly improves the learned representation and the achieved accuracy is on par with the state of the art.

cs.CV↗

arXiv2018

Singular dynamics in the failure of soft adhesive contacts

We characterize the mechanical recovery of compliant silicone gels following adhesive contact failure. We establish broad, stable adhesive contacts between rigid microspheres and soft gels, then stretch the gels to large deformations by pulling quasi-statically on the contact. Eventually, the adhesive contact begins to fail, and ultimately slides to a final contact point on the bottom of the sphere. Immediately after detachment, the gel recoils quickly with a self-similar surface profile that evolves as a power law in time, suggesting that the adhesive detachment point is singular. The singular dynamics we observe are consistent with a relaxation process driven by surface stress and slowed by viscous flow through the porous, elastic network of the gel. Our results emphasize the importance of accounting for both the liquid and solid phases of gels in understanding their mechanics, especially under extreme deformation.

cond-mat.soft↗

arXiv2018

The mechanical equilibrium of soft solids with surface elasticity

Recent experiments have shown that surface stresses in soft materials can have a significant strain-dependence. Here we explore the implications of this surface elasticity to show how, and when, we expect it to arise. We develop the appropriate boundary condition, showing that it simplifies significantly in certain cases. We show that surface elasticity's main role is to effectively stiffen a solid surface's response to in-plane tractions, in particular at length-scales smaller than a characteristic elastocapillary length. We also investigate how surface elasticity effects the Green's-function problem of a line force on a flat, linear-elastic substrate. There are significant changes to this solution, especially in that the well-known displacement singularity is regularised. This raises interesting implications for soft phenomena like wetting contact lines, adhesion and friction. Finally, we discuss open questions, future directions, and close ties with existing fields of research.

cond-mat.soft↗

arXiv2017

Surface Elastic Constants of a Soft Solid

Solid interfaces have intrinsic elasticity. However, in most experiments, this is obscured by bulk stresses. Through microscopic observations of the contact-line geometry of a partially wetting droplet on an anisotropically stretched substrate, we measure two surface-elastic constants that quantify the linear dependence of the surface stress of a soft polymer gel on its strain. With these two parameters, one can predict surface stresses for general deformations of the material in the linear-elastic limit.

cond-mat.soft↗

arXiv2017

Liquid-Liquid Phase Separation in an Elastic Network

Living and engineered systems rely on the stable coexistence of two interspersed liquid phases. Yet surface tension drives their complete separation. Here we show that stable droplets of uniform and tuneable size can be produced through arrested phase separation in an elastic matrix. Starting with an elastic polymer network swollen by a solvent mixture, we change the temperature or composition to drive demixing. Droplets nucleate and grow to a stable size that is tuneable by the network cross-linking density, the cooling rate, and the composition of the solvent mixture. We discuss thermodynamic and mechanical constraints on the process. In particular, we show that the threshold for macroscopic phase separation is altered by the elasticity of the polymer network, and we highlight the role of internuclear correlations in determining the droplet size and polydispersity. This phenomenon has potential applications ranging from colloid synthesis and structural colour to phase separation in biological cells.

cond-mat.soft↗

arXiv2017

Strain-dependent solid surface stress and the stiffness of soft contacts

Surface stresses have recently emerged as a key player in the mechanics of highly compliant solids. The classic theories of contact mechanics describe adhesion with a compliant substrate as a competition between surface energies driving deformation to establish contact and bulk elasticity resisting this. However, it has recently been shown that surface stresses provide an additional restoring force that can compete with and even dominate over elasticity in highly compliant materials, especially when length scales are small compared to the ratio of the surface stress to the elastic modulus, $Υ/E$. Here, we investigate experimentally the contribution of surface stresses to the force of adhesion. We find that the elastic and capillary contributions to the adhesive force are of similar magnitude, and that both are required to account for measured adhesive forces between rigid silica spheres and compliant, silicone gels. Notably, the strain-dependence of the solid surface stress contributes significantly to the stiffness of soft solid contacts.

cond-mat.soft↗

arXiv2017

Direct Measurement of Strain-dependent Solid Surface Stress

Surface stress, also known as surface tension, is a fundamental material property of any interface. However, measurements of solid surface stress in traditional engineering materials, such as metals and oxides, have proven to be very challenging. Consequently, our understanding relies heavily on untested theories, especially regarding the strain dependence of this property. Here, we take advantage of the high compliance and large elastic deformability of a soft polymer gel to directly measure solid surface stress as a function of strain. As anticipated by theoretical work for metals, we find that the surface stress depends on the strain via a surface modulus. Remarkably, the surface modulus of our soft gels is many times larger than the zero-strain surface tension. This suggests that surfaces stresses can play a dominant role in solid mechanics at much larger length scales than previously anticipated.

cond-mat.mtrl-sci↗