Searcharxiv⌕ Search

arXiv subjects

Ji Li

Publications and source records attributed to Ji Li.

At least 109 records · Page 6Linked to original sources
arXiv2024

FontStudio: Shape-Adaptive Diffusion Model for Coherent and Consistent Font Effect Generation

Recently, the application of modern diffusion-based text-to-image generation models for creating artistic fonts, traditionally the domain of professional designers, has garnered significant interest. Diverging from the majority of existing studies that concentrate on generating artistic typography, our research aims to tackle a novel and more demanding challenge: the generation of text effects for multilingual fonts. This task essentially requires generating coherent and consistent visual content within the confines of a font-shaped canvas, as opposed to a traditional rectangular canvas. To address this task, we introduce a novel shape-adaptive diffusion model capable of interpreting the given shape and strategically planning pixel distributions within the irregular canvas. To achieve this, we curate a high-quality shape-adaptive image-text dataset and incorporate the segmentation mask as a visual condition to steer the image generation process within the irregular-canvas. This approach enables the traditionally rectangle canvas-based diffusion model to produce the desired concepts in accordance with the provided geometric shapes. Second, to maintain consistency across multiple letters, we also present a training-free, shape-adaptive effect transfer method for transferring textures from a generated reference letter to others. The key insights are building a font effect noise prior and propagating the font effect information in a concatenated latent space. The efficacy of our FontStudio system is confirmed through user preference studies, which show a marked preference (78% win-rates on aesthetics) for our system even when compared to the latest unrivaled commercial product, Adobe Firefly.

cs.CV↗
arXiv2024

Fractional Bloom boundedness of commutators in spaces of homogeneous type

We aim to characterise boundedness of commutators $[b,T]$ of singular integrals $T$. Boundedness is studied between weighted Lebesgue spaces $L^p(X)$ and $L^q(X)$, $p\leq q$, when the underlying space $X$ is a space of homogeneous type. Commutator theory in spaces of homogeneous type already exist in literature, in particular boundedness results in the setting $p=q$. The purpose here is to extend the earlier results to the setting of $p< q$. Our methods extend those of Duong et al. and Hytönen et al. A novelty here is that in order to show the lower bound of the commutator norm, we demonstrate that the approximate weak factorisation of Hytönen can be used when the underlying setting is a space of homogeneous type and not only in the Euclidean setting. The strength of the approximate weak factorisation is that (when compared to the so-called median method) it readily allows complex-valued $b$ in addition to real-valued ones. However, the median method has been previously successfully applied to iterated commutators and thus has its own strengths. We also present a proof based on that method.

math.CA↗
arXiv2024

Singular Integrals associated with Reflection Groups on Euclidean Space

In the field of harmonic analysis, geometric considerations are frequently crucial. Specially, group actions such as translations, dilations and rotations on Euclidean space are instrumental. The objective of this paper is to extend the study of singular integrals to include the effects of group reflections on Euclidean space, and to establish the T1 theorem for these singular integrals.

math.CA↗
arXiv2024

Global-in-time maximal regularity for the Cauchy problem of the heat equation in BMO and applications

In this article, we establish global-in-time maximal regularity for the Cauchy problem of the classical heat equation $\partial_t u(x,t)-Δu(x,t)=f(x,t)$ with $u(x,0)=0$ in a certain $\rm BMO$ setting, which improves the local-in-time result initially proposed by Ogawa and Shimizu in \cite{OS, OS2}. In further developing our method originally formulated for the heat equation, we obtain analogous global ${\rm BMO}$-maximal regularity associated to the Schrödinger operator $\mathcal L=-Δ+V$, where the nonnegative potential $V$ belongs to the reverse Hölder class ${\rm RH}_q$ for some $q> n/2$. This extension includes several inhomogeneous estimates as ingredients, such as Carleson-type estimates for the external forces. Our new methodology is to exploit elaborate heat kernel estimates, along with matched space-time decomposition on the involving integral-type structure of maximal operators, as well as some global techniques such as those from de Simon's work and Schur's lemma. One crucial trick is to utilize the mean oscillation therein to contribute a higher and necessary decay order for global-in-time estimates.

math.AP↗
arXiv2024

Muckenhoupt-Type Weights and Quantitative Weighted Estimate in the Bessel Setting

Part of the intrinsic structure of singular integrals in the Bessel setting is captured by Muckenhoupt-type weights. Anderson--Kerman showed that the Bessel Riesz transform is bounded on weighted $L^p_w$ if and only if $w$ is in the class $A_{p,λ}$. We introduce a new class of Muckenhoupt-type weights $\widetilde A_{p,λ}$ in the Bessel setting, which is different from $A_{p,λ}$ but characterizes the weighted boundedness for the Hardy--Littlewood maximal operators. We also establish the weighted $L^p$ boundedness and compactness, as well as the endpoint weak type boundedness of Riesz commutators. The quantitative weighted bound is also established.

math.CA↗
arXiv2024

The stability of smooth solitary waves for the $b$-family of Camassa-Holm equations

The $b$-family of Camassa-Holm ($b$-CH) equation is a one-parameter family of PDEs, which includes the completely integrable Camassa-Holm and Degasperis-Procesi equations but possesses different Hamiltonian structures. Motivated by this, we study the existence and the orbital stability of the smooth solitary wave solutions with nonzero constant background to the $b$-CH equation for the special case $b=1$, whose the Hamiltonian structure is different from that of $b\neq1$. We establish a connection between the stability criterion for the solitary waves and the monotonicity of a singular integral along the corresponding homoclinic orbits of the spatial ODEs. We verify the latter analytically using the framework for the monotonicity of period function of planar Hamiltonian systems, which shows that the smooth solitary waves are orbitally stable. In addition, we find that the existence and orbital stability results for $0 1$, particularly the stability criteria are the same. Finally, combining with the results for the case $b>1$, we conclude that the solitary waves to the $b$-CH equation is structurally stable under the variation of %with respect to the parameter $b>0$.

math.AP↗
arXiv2024

DesignEdit: Multi-Layered Latent Decomposition and Fusion for Unified & Accurate Image Editing

Recently, how to achieve precise image editing has attracted increasing attention, especially given the remarkable success of text-to-image generation models. To unify various spatial-aware image editing abilities into one framework, we adopt the concept of layers from the design domain to manipulate objects flexibly with various operations. The key insight is to transform the spatial-aware image editing task into a combination of two sub-tasks: multi-layered latent decomposition and multi-layered latent fusion. First, we segment the latent representations of the source images into multiple layers, which include several object layers and one incomplete background layer that necessitates reliable inpainting. To avoid extra tuning, we further explore the inner inpainting ability within the self-attention mechanism. We introduce a key-masking self-attention scheme that can propagate the surrounding context information into the masked region while mitigating its impact on the regions outside the mask. Second, we propose an instruction-guided latent fusion that pastes the multi-layered latent representations onto a canvas latent. We also introduce an artifact suppression scheme in the latent space to enhance the inpainting quality. Due to the inherent modular advantages of such multi-layered representations, we can achieve accurate image editing, and we demonstrate that our approach consistently surpasses the latest spatial editing methods, including Self-Guidance and DiffEditor. Last, we show that our approach is a unified framework that supports various accurate image editing tasks on more than six different editing tasks.

cs.CV↗
arXiv2024

COLE: A Hierarchical Generation Framework for Multi-Layered and Editable Graphic Design

Graphic design, which has been evolving since the 15th century, plays a crucial role in advertising. The creation of high-quality designs demands design-oriented planning, reasoning, and layer-wise generation. Unlike the recent CanvaGPT, which integrates GPT-4 with existing design templates to build a custom GPT, this paper introduces the COLE system - a hierarchical generation framework designed to comprehensively address these challenges. This COLE system can transform a vague intention prompt into a high-quality multi-layered graphic design, while also supporting flexible editing based on user input. Examples of such input might include directives like ``design a poster for Hisaishi's concert.'' The key insight is to dissect the complex task of text-to-design generation into a hierarchy of simpler sub-tasks, each addressed by specialized models working collaboratively. The results from these models are then consolidated to produce a cohesive final output. Our hierarchical task decomposition can streamline the complex process and significantly enhance generation reliability. Our COLE system comprises multiple fine-tuned Large Language Models (LLMs), Large Multimodal Models (LMMs), and Diffusion Models (DMs), each specifically tailored for design-aware layer-wise captioning, layout planning, reasoning, and the task of generating images and text. Furthermore, we construct the DESIGNINTENTION benchmark to demonstrate the superiority of our COLE system over existing methods in generating high-quality graphic designs from user intent. Last, we present a Canva-like multi-layered image editing tool to support flexible editing of the generated multi-layered graphic design images. We perceive our COLE system as an important step towards addressing more complex and multi-layered graphic design generation tasks in the future.

cs.CV↗
arXiv2024

Schatten Properties of Calderón--Zygmund Singular Integral Commutator on stratified Lie groups

We provide full characterisation of the Schatten properties of $[M_b,T]$, the commutator of Calderón--Zygmund singular integral $T$ with symbol $b$ $(M_bf(x):=b(x)f(x))$ on stratified Lie groups $\mathbb{G}$. We show that, when $p$ is larger than the homogeneous dimension $\mathbb{Q}$ of $\mathbb{G}$, the Schatten $\mathcal{L}_p$ norm of the commutator is equivalent to the Besov semi-norm $B_{p}^{\frac{\mathbb{Q}}{p}}$ of the function $b$; but when $p\leq \mathbb{Q}$, the commutator belongs to $\mathcal{L}_p$ if and only if $b$ is a constant. For the endpoint case at the critical index $p=\mathbb{Q}$, we further show that the Schatten $\mathcal{L}_{\mathbb{Q},\infty}$ norm of the commutator is equivalent to the Sobolev norm $W^{1,\mathbb{Q}}$ of $b$. Our method at the endpoint case differs from existing methods of Fourier transforms or trace formula for Euclidean spaces or Heisenberg groups, respectively, and hence can be applied to various settings beyond.

math.CA↗
arXiv2024

Schatten--Lorentz characterization of Riesz transform commutator associated with Bessel operators

Let $Δ_λ$ be the Bessel operator on the upper half space $\mathbb{R}_+^{n+1}$ with $n\geq 0$ and $λ>0$, and $R_{λ,j}$ be the $j-$th Bessel Riesz transform, $j=1,\ldots,n+1$. We demonstrate that the Schatten--Lorentz norm ($S^{p,q}$, $1<p<\infty$, $1\leq q\leq \infty$) of the commutator $[b,R_{λ,j}]$ can be characterized in terms of the oscillation space norm of the symbol $b$. In particular, for the case $p=q$, the Schatten norm of $[b,R_{λ,j}]$ can be further characterized in terms of the Besov norm of the symbol. Moreover, the critical index is also studied, which is $p=n+1$, the lower dimension of the Bessel measure (but not the upper dimension). Our approach relies on martingale and dyadic analysis, which enables us to bypass the use of Fourier analysis effectively.

math.FA↗
arXiv2024

Revisiting Disentanglement in Downstream Tasks: A Study on Its Necessity for Abstract Visual Reasoning

In representation learning, a disentangled representation is highly desirable as it encodes generative factors of data in a separable and compact pattern. Researchers have advocated leveraging disentangled representations to complete downstream tasks with encouraging empirical evidence. This paper further investigates the necessity of disentangled representation in downstream applications. Specifically, we show that dimension-wise disentangled representations are unnecessary on a fundamental downstream task, abstract visual reasoning. We provide extensive empirical evidence against the necessity of disentanglement, covering multiple datasets, representation learning methods, and downstream network architectures. Furthermore, our findings suggest that the informativeness of representations is a better indicator of downstream performance than disentanglement. Finally, the positive correlation between informativeness and disentanglement explains the claimed usefulness of disentangled representations in previous works. The source code is available at https://github.com/Richard-coder-Nai/disentanglement-lib-necessity.git.

cs.CV↗
arXiv2024

Riesz transform and commutators in the Dunkl setting

In this paper we characterise the optimal pointwise size and regularity estimates for the Dunkl Riesz transform kernel involving both the Euclidean metric and the Dunkl metric, where these two metrics are not equivalent. We further establish a suitable version of the pointwise kernel lower bound of the Dunkl Riesz transform via the Euclidean metric only. Then we show that the lower bound of commutator of the Dunkl Riesz transform is with respect to the BMO space associated with the Euclidean metric, and that the upper bound is respect to the BMO space associated with the Dunkl metric. Moreover, the compactness and the two types of VMO are also addressed.

math.CA↗
arXiv2023

Privacy-Aware Energy Consumption Modeling of Connected Battery Electric Vehicles using Federated Learning

Battery Electric Vehicles (BEVs) are increasingly significant in modern cities due to their potential to reduce air pollution. Precise and real-time estimation of energy consumption for them is imperative for effective itinerary planning and optimizing vehicle systems, which can reduce driving range anxiety and decrease energy costs. As public awareness of data privacy increases, adopting approaches that safeguard data privacy in the context of BEV energy consumption modeling is crucial. Federated Learning (FL) is a promising solution mitigating the risk of exposing sensitive information to third parties by allowing local data to remain on devices and only sharing model updates with a central server. Our work investigates the potential of using FL methods, such as FedAvg, and FedPer, to improve BEV energy consumption prediction while maintaining user privacy. We conducted experiments using data from 10 BEVs under simulated real-world driving conditions. Our results demonstrate that the FedAvg-LSTM model achieved a reduction of up to 67.84\% in the MAE value of the prediction results. Furthermore, we explored various real-world scenarios and discussed how FL methods can be employed in those cases. Our findings show that FL methods can effectively improve the performance of BEV energy consumption prediction while maintaining user privacy.

cs.LG↗
arXiv2023

Muckenhoupt-type weights and the intrinsic structure in Bessel Setting

Fix $λ>-1/2$ and $λ\not=0$. Consider the Bessel operator (introduced by Muckenhoupt--Stein) $\triangle_λ:=-\frac{d^2}{dx^2}-\frac{2λ}{x} \frac d{dx}$ on $\mathbb{R_+}:=(0,\infty)$ with $dm_λ(x):=x^{2λ}dx$ and $dx$ the Lebesgue measure on $\mathbb{R_+}$. In this paper, we study the Muckenhoupt-type weights which reveal the intrinsic structure in this Bessel setting along the line of Muckenhoupt--Stein and Andersen--Kerman. Besides, exploiting more properties of the weights $A_{p,λ}$ introduced by Andersen--Kerman, we introduce a new class $\widetilde{A}_{p,λ}$ such that the Hardy--Littlewood maximal function is bounded on the weighted $L^p_w$ space if and only if $w$ is in $\widetilde A_{p,λ}$. Moreover, along the line of Coifman--Rochberg--Weiss, we investigate the commutator $[b,R_λ]$ with $R_λ:=\frac{d}{dx}(\triangle_λ)^{-\frac{1}{2}}$ to be the Bessel Riesz transform. We show that for $w\in A_{p,λ}$, the commutator $[b, R_λ]$ is bounded on weighted $L^p_w$ if and only if $b$ is in the BMO space associated with $\triangle_λ$.

math.CA↗
arXiv2023

On Fefferman--Stein type inequality on Shilov boundaries and applications

In this paper, we establish the Fefferman--Stein type inequality for area integral and non-tangential maximal function on the Shilov boundary studied by Nagel and Stein in 2004. The technique here is inspired by Fefferman--Stein (1972) and Merryfield (1985) but we bypass the use of Fourier or group structure as these were not available on the polynomial domains of finite type. Direct applications include the maximal function characterisation of product Hardy space and the weak type endpoint estimate for product Calderón--Zygmund operators (such as the Cauchy--Szegő projection) on the Shilov boundary.

math.CV↗
arXiv2023

Characterizations of product Hardy spaces on stratified groups by singular integrals and maximal functions

A large part of the theory of Hardy spaces on products of Euclidean spaces has been extended to the setting of products of stratified Lie groups. This includes characterisation of Hardy spaces by square functions and by atomic decompositions, proof of the duality of Hardy spaces with BMO, and description of many interpolation spaces. Until now, however, two aspects of the classical theory have been conspicuously absent: the characterisation of Hardy spaces by singular integrals (of Christ--Geller type) or by (vertical or nontangential) maximal functions. In this paper we fill in these gaps by developing new techniques on products of stratified groups, using the ideas of Chen, Cowling, Lee, Li and Ottazzi on the Heisenberg group with flag structure.

math.FA↗
arXiv2023

A Review on AI Algorithms for Energy Management in E-Mobility Services

E-mobility, or electric mobility, has emerged as a pivotal solution to address pressing environmental and sustainability concerns in the transportation sector. The depletion of fossil fuels, escalating greenhouse gas emissions, and the imperative to combat climate change underscore the significance of transitioning to electric vehicles (EVs). This paper seeks to explore the potential of artificial intelligence (AI) in addressing various challenges related to effective energy management in e-mobility systems (EMS). These challenges encompass critical factors such as range anxiety, charge rate optimization, and the longevity of energy storage in EVs. By analyzing existing literature, we delve into the role that AI can play in tackling these challenges and enabling efficient energy management in EMS. Our objectives are twofold: to provide an overview of the current state-of-the-art in this research domain and propose effective avenues for future investigations. Through this analysis, we aim to contribute to the advancement of sustainable and efficient e-mobility solutions, shaping a greener and more sustainable future for transportation.

cs.LG↗