arXiv subjects

Xuping Xie

Publications and source records attributed to Xuping Xie.

13 recordsLinked to original sources

arXiv2025

OmniStyle2: Learning to Stylize by Learning to Destylize

This paper introduces a scalable paradigm for supervised style transfer by inverting the problem: instead of learning to stylize directly, we learn to destylize, reducing stylistic elements from artistic images to recover their natural counterparts and thereby producing authentic, pixel-aligned training pairs at scale. To realize this paradigm, we propose DeStylePipe, a progressive, multi-stage destylization framework that begins with global general destylization, advances to category-wise instruction adaptation, and ultimately deploys specialized model adaptation for complex styles that prompt engineering alone cannot handle. Tightly integrated into this pipeline, DestyleCoT-Filter employs Chain-of-Thought reasoning to assess content preservation and style removal at each stage, routing challenging samples forward while discarding persistently low-quality pairs. Built on this framework, we construct DeStyle-350K, a large-scale dataset aligning diverse artistic styles with their underlying content. We further introduce BCS-Bench, a benchmark featuring balanced content generality and style diversity for systematic evaluation. Extensive experiments demonstrate that models trained on DeStyle-350K achieve superior stylization quality, validating destylization as a reliable and scalable supervision paradigm for style transfer.

cs.CV↗

arXiv2025

SigStyle: Signature Style Transfer via Personalized Text-to-Image Models

Style transfer enables the seamless integration of artistic styles from a style image into a content image, resulting in visually striking and aesthetically enriched outputs. Despite numerous advances in this field, existing methods did not explicitly focus on the signature style, which represents the distinct and recognizable visual traits of the image such as geometric and structural patterns, color palettes and brush strokes etc. In this paper, we introduce SigStyle, a framework that leverages the semantic priors that embedded in a personalized text-to-image diffusion model to capture the signature style representation. This style capture process is powered by a hypernetwork that efficiently fine-tunes the diffusion model for any given single style image. Style transfer then is conceptualized as the reconstruction process of content image through learned style tokens from the personalized diffusion model. Additionally, to ensure the content consistency throughout the style transfer process, we introduce a time-aware attention swapping technique that incorporates content information from the original image into the early denoising steps of target image generation. Beyond enabling high-quality signature style transfer across a wide range of styles, SigStyle supports multiple interesting applications, such as local style transfer, texture transfer, style fusion and style-guided text-to-image generation. Quantitative and qualitative evaluations demonstrate our approach outperforms existing style transfer methods for recognizing and transferring the signature styles.

cs.GR↗

arXiv2025

DP-Adapter: Dual-Pathway Adapter for Boosting Fidelity and Text Consistency in Customizable Human Image Generation

With the growing popularity of personalized human content creation and sharing, there is a rising demand for advanced techniques in customized human image generation. However, current methods struggle to simultaneously maintain the fidelity of human identity and ensure the consistency of textual prompts, often resulting in suboptimal outcomes. This shortcoming is primarily due to the lack of effective constraints during the simultaneous integration of visual and textual prompts, leading to unhealthy mutual interference that compromises the full expression of both types of input. Building on prior research that suggests visual and textual conditions influence different regions of an image in distinct ways, we introduce a novel Dual-Pathway Adapter (DP-Adapter) to enhance both high-fidelity identity preservation and textual consistency in personalized human image generation. Our approach begins by decoupling the target human image into visually sensitive and text-sensitive regions. For visually sensitive regions, DP-Adapter employs an Identity-Enhancing Adapter (IEA) to preserve detailed identity features. For text-sensitive regions, we introduce a Textual-Consistency Adapter (TCA) to minimize visual interference and ensure the consistency of textual semantics. To seamlessly integrate these pathways, we develop a Fine-Grained Feature-Level Blending (FFB) module that efficiently combines hierarchical semantic features from both pathways, resulting in more natural and coherent synthesis outcomes. Additionally, DP-Adapter supports various innovative applications, including controllable headshot-to-full-body portrait generation, age editing, old-photo to reality, and expression editing.

cs.GR↗

arXiv2024

Latent Space Dynamics Learning for Stiff Collisional-radiative Models

In this work, we propose a data-driven method to discover the latent space and learn the corresponding latent dynamics for a collisional-radiative (CR) model in radiative plasma simulations. The CR model, consisting of high-dimensional stiff ordinary differential equations (ODEs), must be solved at each grid point in the configuration space, leading to significant computational costs in plasma simulations. Our method employs a physics-assisted autoencoder to extract a low-dimensional latent representation of the original CR system. A flow map neural network is then used to learn the latent dynamics. Once trained, the reduced surrogate model predicts the entire latent dynamics given only the initial condition by iteratively applying the flow map. The radiative power loss is then reconstructed using a decoder. Numerical experiments demonstrate that the proposed architecture can accurately predict both the full-order CR dynamics and the radiative power loss rate.

physics.plasm-ph↗

arXiv2020

Reduced Order Models for the Quasi-Geostrophic Equations: A Brief Survey

Reduced order models (ROMs) are computational models whose dimension is significantly lower than those obtained through classical numerical discretizations (e.g., finite element, finite difference, finite volume, or spectral methods). Thus, ROMs have been used to accelerate numerical simulations of many query problems, e.g., uncertainty quantification, control, and shape optimization. Projection-based ROMs have been particularly successful in the numerical simulation of fluid flows. In this brief survey, we summarize some recent ROM developments for the quasi-geostrophic equations (QGE) (also known as the barotropic vorticity equations), which are a simplified model for geophysical flows in which rotation plays a central role, such as wind-driven ocean circulation in mid-latitude ocean basins. Since the QGE represent a practical compromise between efficient numerical simulations of ocean flows and accurate representations of large scale ocean dynamics, these equations have often been used in the testing of new numerical methods for ocean flows. ROMs have also been tested on the QGE for various settings in order to understand their potential in efficient numerical simulations of ocean flows. In this paper, we survey the ROMs developed for the QGE in order to understand their potential in efficient numerical simulations of more complex ocean flows: We explain how classical numerical methods for the QGE are used to generate the ROM basis functions, we outline the main steps in the construction of projection-based ROMs (with a particular focus on the under-resolved regime, when the closure problem needs to be addressed), we illustrate the ROMs in the numerical simulation of the QGE for various settings, and we present several potential future research avenues in the ROM exploration of the QGE and more complex models of geophysical flows.

physics.flu-dyn↗

arXiv2020

Lagrangian Data-Driven Reduced Order Modeling of Finite Time Lyapunov Exponents

There are two main strategies for improving the projection-based reduced order model (ROM) accuracy: (i) improving the ROM, i.e., adding new terms to the standard ROM; and (ii) improving the ROM basis, i.e., constructing ROM bases that yield more accurate ROMs. In this paper, we use the latter. We propose new Lagrangian inner products that we use together with Eulerian and Lagrangian data to construct new Lagrangian ROMs. We show that the new Lagrangian ROMs are orders of magnitude more accurate than the standard Eulerian ROMs, i.e., ROMs that use standard Eulerian inner product and data to construct the ROM basis. Specifically, for the quasi-geostrophic equations, we show that the new Lagrangian ROMs are more accurate than the standard Eulerian ROMs in approximating not only Lagrangian fields (e.g., the finite time Lyapunov exponent (FTLE)), but also Eulerian fields (e.g., the streamfunction). We emphasize that the new Lagrangian ROMs do not employ any closure modeling to model the effect of discarded modes (which is standard procedure for low-dimensional ROMs of complex nonlinear systems). Thus, the dramatic increase in the new Lagrangian ROMs' accuracy is entirely due to the novel Lagrangian inner products used to build the Lagrangian ROM basis.

physics.flu-dyn↗

arXiv2019

Closure Learning for Nonlinear Model Reduction Using Deep Residual Neural Network

Developing accurate, efficient, and robust closure models is essential in the construction of reduced order models (ROMs) for realistic nonlinear systems, which generally require drastic ROM mode truncations. We propose a deep residual neural network (ResNet) closure learning framework for ROMs of nonlinear systems. The novel ResNet-ROM framework consists of two steps: (i) In the first step, we use ROM projection to filter the given nonlinear PDE and construct a filtered ROM. This filtered ROM is low-dimensional, but is not closed (because of the PDE nonlinearity). (ii) In the second step, we use ResNet to close the filtered ROM, i.e., to model the interaction between the resolved and unresolved ROM modes. We emphasize that in the new ResNet-ROM framework, data is used only to complement classical physical modeling (i.e., only in the closure modeling component), not to completely replace it. We also note that the new ResNet-ROM is built on general ideas of spatial filtering and deep learning and is independent of (restrictive) phenomenological arguments, e.g., of eddy viscosity type. The numerical experiments for the 1D Burgers equation show that the ResNet-ROM is significantly more accurate than the standard projection ROM. The new ResNet-ROM is also more accurate and significantly more efficient than other modern ROM closure models.

physics.flu-dyn↗

arXiv2019

Analytic Continuation of Noisy Data Using Adams Bashforth ResNet

We propose a data-driven learning framework for the analytic continuation problem in numerical quantum many-body physics. Designing an accurate and efficient framework for the analytic continuation of imaginary time using computational data is a grand challenge that has hindered meaningful links with experimental data. The standard Maximum Entropy (MaxEnt)-based method is limited by the quality of the computational data and the availability of prior information. Also, the MaxEnt is not able to solve the inversion problem under high level of noise in the data. Here we introduce a novel learning model for the analytic continuation problem using a Adams-Bashforth residual neural network (AB-ResNet). The advantage of this deep learning network is that it is model independent and, therefore, does not require prior information concerning the quantity of interest given by the spectral function. More importantly, the ResNet-based model achieves higher accuracy than MaxEnt for data with higher level of noise. Finally, numerical examples show that the developed AB-ResNet is able to recover the spectral function with accuracy comparable to MaxEnt where the noise level is relatively small.

physics.comp-ph↗

arXiv2018

Non-intrusive inference reduced order model for fluids using linear multistep neural network

In this effort we propose a data-driven learning framework for reduced order modeling of fluid dynamics. Designing accurate and efficient reduced order models for nonlinear fluid dynamic problems is challenging for many practical engineering applications. Classical projection-based model reduction methods generate reduced systems by projecting full-order differential operators into low-dimensional subspaces. However, these techniques usually lead to severe instabilities in the presence of highly nonlinear dynamics, which dramatically deteriorates the accuracy of the reduced-order models. In contrast, our new framework exploits linear multistep networks, based on implicit Adams-Moulton schemes, to construct the reduced system. The advantage is that the method optimally approximates the full order model in the low-dimensional space with a given supervised learning task. Moreover, our approach is non-intrusive, such that it can be applied to other complex nonlinear dynamical systems with sophisticated legacy codes. We demonstrate the performance of our method through the numerical simulation of a two-dimensional flow past a circular cylinder with Reynolds number Re = 100. The results reveal that the new data-driven model is significantly more accurate than standard projection-based approaches.

physics.comp-ph↗

arXiv2017

Numerical Analysis of the Leray Reduced Order Model

Standard ROMs generally yield spurious numerical oscillations in the simulation of convection-dominated flows. Regularized ROMs use explicit ROM spatial filtering to decrease these spurious numerical oscillations. The Leray ROM is a recently introduced regularized ROM that utilizes explicit ROM spatial filtering of the convective term in the Navier-Stokes equations. This paper presents the numerical analysis of the finite element discretization of the Leray ROM. Error estimates for the ROM differential filter, which is the explicit ROM spatial filter used in the Leray ROM, are proved. These ROM filtering error estimates are then used to prove error estimates for the Leray ROM. Finally, both the ROM filtering error estimates and the Leray ROM error estimates are numerically investigated in the simulation of the two-dimensional Navier-Stokes equations with an analytic solution.

math.NA↗

arXiv2017

Regularized Reduced Order Models for a Stochastic Burgers Equation

In this paper, we study the numerical stability of reduced order models for convection-dominated stochastic systems in a relatively simple setting: a stochastic Burgers equation with linear multiplicative noise. Our preliminary results suggest that, in a convection-dominated regime, standard reduced order models yield inaccurate results in the form of spurious numerical oscillations. To alleviate these oscillations, we use the Leray reduced order model, which increases the numerical stability of the standard model by smoothing (regularizing) the convective term with an explicit spatial filter. The Leray reduced order model yields significantly better results than the standard reduced order model and is more robust with respect to changes in the strength of the noise.

physics.flu-dyn↗

arXiv2016

An Evolve-Then-Filter Regularized Reduced Order Model

In this paper, we propose a new evolve-then-filter reduced order model (EF-ROM). This is a regularized ROM (Reg-ROM), which aims at the numerical stabilization of proper orthogonal decomposition (POD) ROMs for convection-dominated flows. We also consider the Leray ROM (L-ROM). These two Reg-ROMs use explicit ROM spatial filtering to smooth (regularize) various terms in the ROMs. Two spatial filters are used: a POD projection onto a POD subspace (Proj) and a new POD differential filter (DF). The four Reg-ROM/filter combinations are tested in the numerical simulation of the three-dimensional flow past a circular cylinder at a Reynolds number $Re=1000$. Overall, the most accurate Reg-ROM/filter combination is EF-ROM-DF. Furthermore, the spatial filter has a higher impact on the Reg-ROM than the regularization used. Indeed, the DF generally yields better results than Proj for both the EF-ROM and L-ROM. Finally, the CPU times of the four Reg-ROM/filter combinations are orders of magnitude lower than the CPU time of the DNS.

physics.flu-dyn↗

arXiv2015

Approximate Deconvolution Reduced Order Modeling

This paper proposes a large eddy simulation reduced order model(LES-ROM) framework for the numerical simulation of realistic flows. In this LES-ROM framework, the proper orthogonal decomposition(POD) is used to define the ROM basis and a POD differential filter is used to define the large ROM structures. An approximate deconvolution(AD) approach is used to solve the ROM closure problem and develop a new AD-ROM. This AD-ROM is tested in the numerical simulation of the one-dimensional Burgers equation with a small diffusion coefficient(10^{-3})

physics.flu-dyn↗