Searcharxiv⌕ Search

arXiv subjects

Wei Sun

Publications and source records attributed to Wei Sun.

At least 271 records · Page 15Linked to original sources
arXiv2023

AGIQA-3K: An Open Database for AI-Generated Image Quality Assessment

With the rapid advancements of the text-to-image generative model, AI-generated images (AGIs) have been widely applied to entertainment, education, social media, etc. However, considering the large quality variance among different AGIs, there is an urgent need for quality models that are consistent with human subjective ratings. To address this issue, we extensively consider various popular AGI models, generated AGI through different prompts and model parameters, and collected subjective scores at the perceptual quality and text-to-image alignment, thus building the most comprehensive AGI subjective quality database AGIQA-3K so far. Furthermore, we conduct a benchmark experiment on this database to evaluate the consistency between the current Image Quality Assessment (IQA) model and human perception, while proposing StairReward that significantly improves the assessment performance of subjective text-to-image alignment. We believe that the fine-grained subjective scores in AGIQA-3K will inspire subsequent AGI quality models to fit human subjective perception mechanisms at both perception and alignment levels and to optimize the generation result of future AGI models. The database is released on https://github.com/lcysyzxdxc/AGIQA-3k-Database.

cs.CV↗
arXiv2023

Learning Prescriptive ReLU Networks

We study the problem of learning optimal policy from a set of discrete treatment options using observational data. We propose a piecewise linear neural network model that can balance strong prescriptive performance and interpretability, which we refer to as the prescriptive ReLU network, or P-ReLU. We show analytically that this model (i) partitions the input space into disjoint polyhedra, where all instances that belong to the same partition receive the same treatment, and (ii) can be converted into an equivalent prescriptive tree with hyperplane splits for interpretability. We demonstrate the flexibility of the P-ReLU network as constraints can be easily incorporated with minor modifications to the architecture. Through experiments, we validate the superior prescriptive accuracy of P-ReLU against competing benchmarks. Lastly, we present examples of interpretable prescriptive trees extracted from trained P-ReLUs using a real-world dataset, for both the unconstrained and constrained scenarios.

cs.LG↗
arXiv2023

Viscosity solutions to uniformly elliptic complex equations

In this paper, we shall extend the definition of $\mathcal{C}$-subsolution condition and adapt the argument of Guo-Phong-Tong[18] to replace Alexandroff-Bakelman-Pucci estimate in complex cases. As an application, we shall define and study the viscosity solutions to uniformly elliptic complex equations and prove the Hölder regularity, following the argument for real equations. Our results show that the new method can improve the dependence in regularity and a priori estimates for complex elliptic equations.

math.AP↗
arXiv2023

$η$-glueball mixing from $N_f=2$ lattice QCD

We perform the first lattice study on the mixing of the isoscalar pseudoscalar meson $η$ and the pseudoscalar glueball $G$ in the $N_f=2$ QCD at the pion mass $m_π\approx 350$ MeV. The $η$ mass is determined to be $m_η=714(6)(16)$ MeV. Through the Witten-Veneziano relation, this value can be matched to a mass value of $\sim 981$ MeV for the $\mathrm{SU(3)}$ counterpart of $η$. Based on a large gauge ensemble, the $η-G$ mixing energy and the mixing angle are determined to be $|x|=107(15)(2)$ MeV and $|θ|=3.46(46)^\circ$ from the $η-G$ correlators that are calculated using the distillation method. We conclude that the $η-G$ mixing is tiny and the topology induced interaction contributes most of $η$ mass owing to the QCD $\mathrm{U_A(1)}$ anomaly.

hep-lat↗
arXiv2023

Variation comparison between the $F$-distribution and the normal distribution

Let $X_{d_1,d_2}$ be an $F$-random variable with numerator and denominator degrees of freedom $d_1$ and $d_2$, respectively. We investigate the inequality: $P\{|X_{d_1,d_2}-E[X_{d_1,d_2}]|\le \sqrt{{\rm Var}(X_{d_1,d_2})}\}\ge P\{|W-E[W]|\le \sqrt{{\rm Var}(W)}\}$, where $W$ is a standard normal random variable or a $χ^2(d_1)$ random variable. We prove that this inequality holds for $d_1\in\{1,2,3,4\}$ and $5\le d_2\in\mathbb{N}$.

math.PR↗
arXiv2023

Variation comparison between infinitely divisible distributions and the normal distribution

Let $X$ be a random variable with finite second moment. We investigate the inequality: $P\{|X-E[X]|\le \sqrt{{\rm Var}(X)}\}\ge P\{|Z|\le 1\}$, where $Z$ is a standard normal random variable. We prove that this inequality holds for many familiar infinitely divisible continuous distributions including the Laplace, Gumbel, Logistic, Pareto, infinitely divisible Weibull, log-normal, student's $t$ and inverse Gaussian distributions. Numerical results are given to show that the inequality with continuity correction also holds for some infinitely divisible discrete distributions.

math.PR↗
arXiv2023

Blind Quality Assessment for in-the-Wild Images via Hierarchical Feature Fusion and Iterative Mixed Database Training

Image quality assessment (IQA) is very important for both end-users and service providers since a high-quality image can significantly improve the user's quality of experience (QoE) and also benefit lots of computer vision algorithms. Most existing blind image quality assessment (BIQA) models were developed for synthetically distorted images, however, they perform poorly on in-the-wild images, which are widely existed in various practical applications. In this paper, we propose a novel BIQA model for in-the-wild images by addressing two critical problems in this field: how to learn better quality-aware feature representation, and how to solve the problem of insufficient training samples in terms of their content and distortion diversity. Considering that perceptual visual quality is affected by both low-level visual features (e.g. distortions) and high-level semantic information (e.g. content), we first propose a staircase structure to hierarchically integrate the features from intermediate layers into the final feature representation, which enables the model to make full use of visual information from low-level to high-level. Then an iterative mixed database training (IMDT) strategy is proposed to train the BIQA model on multiple databases simultaneously, so the model can benefit from the increase in both training samples and image content and distortion diversity and can learn a more general feature representation. Experimental results show that the proposed model outperforms other state-of-the-art BIQA models on six in-the-wild IQA databases by a large margin. Moreover, the proposed model shows an excellent performance in the cross-database evaluation experiments, which further demonstrates that the learned feature representation is robust to images with diverse distortions and content. The code is available at https://github.com/sunwei925/StairIQA.

cs.MM↗
arXiv2023

$Ξ_c-Ξ_c^{\prime}$ mixing From Lattice QCD

In heavy quark limit, the lowest-lying charmed baryons with two light quarks can form an SU(3) triplet and sextet. The $Ξ_c$ in the SU(3) triplet and $Ξ_c'$ in the sextet have the same $J^{PC}$ quantum number and can mix due to the finite charm quark mass and the fact the strange quark is heavier than the up/down quark. We explore the $Ξ_c$-$Ξ_c'$ mixing by calculating the two-point correlation functions of the $Ξ_c$ and $Ξ_c'$ baryons from lattice QCD. Based on the lattice data, we adopt two independent methods to determine the mixing angle between $Ξ_c$ and $Ξ_c'$. After making the chiral and continuum extrapolation, it is found that the mixing angle $θ$ is $1.2^{\circ}\pm0.1^{\circ}$, which seems insufficient to account for the large SU(3) symmetry breaking effects found in weak decays of charmed baryons.

hep-lat↗
arXiv2023

MM-PCQA: Multi-Modal Learning for No-reference Point Cloud Quality Assessment

The visual quality of point clouds has been greatly emphasized since the ever-increasing 3D vision applications are expected to provide cost-effective and high-quality experiences for users. Looking back on the development of point cloud quality assessment (PCQA) methods, the visual quality is usually evaluated by utilizing single-modal information, i.e., either extracted from the 2D projections or 3D point cloud. The 2D projections contain rich texture and semantic information but are highly dependent on viewpoints, while the 3D point clouds are more sensitive to geometry distortions and invariant to viewpoints. Therefore, to leverage the advantages of both point cloud and projected image modalities, we propose a novel no-reference point cloud quality assessment (NR-PCQA) metric in a multi-modal fashion. In specific, we split the point clouds into sub-models to represent local geometry distortions such as point shift and down-sampling. Then we render the point clouds into 2D image projections for texture feature extraction. To achieve the goals, the sub-models and projected images are encoded with point-based and image-based neural networks. Finally, symmetric cross-modal attention is employed to fuse multi-modal quality-aware information. Experimental results show that our approach outperforms all compared state-of-the-art methods and is far ahead of previous NR-PCQA methods, which highlights the effectiveness of the proposed method. The code is available at https://github.com/zzc-1998/MM-PCQA.

cs.CV↗
arXiv2023

MD-VQA: Multi-Dimensional Quality Assessment for UGC Live Videos

User-generated content (UGC) live videos are often bothered by various distortions during capture procedures and thus exhibit diverse visual qualities. Such source videos are further compressed and transcoded by media server providers before being distributed to end-users. Because of the flourishing of UGC live videos, effective video quality assessment (VQA) tools are needed to monitor and perceptually optimize live streaming videos in the distributing process. In this paper, we address \textbf{UGC Live VQA} problems by constructing a first-of-a-kind subjective UGC Live VQA database and developing an effective evaluation tool. Concretely, 418 source UGC videos are collected in real live streaming scenarios and 3,762 compressed ones at different bit rates are generated for the subsequent subjective VQA experiments. Based on the built database, we develop a \underline{M}ulti-\underline{D}imensional \underline{VQA} (\textbf{MD-VQA}) evaluator to measure the visual quality of UGC live videos from semantic, distortion, and motion aspects respectively. Extensive experimental results show that MD-VQA achieves state-of-the-art performance on both our UGC Live VQA database and existing compressed UGC VQA databases.

cs.CV↗
arXiv2023

$1^{-+}$ Hybrid in $J/ψ$ Radiative Decays from Lattice QCD

We present the first theoretical prediction of the production rate of $1^{-+}$ light hybrid meson $η_1$ in $J/ψ$ radiative decays. In the $N_f=2$ lattice QCD formalism with the pion mass $m_π\approx 350$ MeV, the related electromagnetic multipole form factors are extracted from the three-point functions that involve necessarily quark annihilation diagrams, which are calculated through the distillation method. The partial width of $J/ψ\to γη_1$ is determined to be $2.29(77)~\mathrm{eV}$ at the $η_1$ mass $m_{η_1}=2.23(4)$ GeV. If $η_1$ corresponds to the recently observed $η_1(1855)$ in the process $J/ψ\to γη_1(1855)\to γηη'$ by BESIII, then the branching fraction $\mathrm{Br}(J/ψ\to γη_1(1855))$ is estimated to be $6.2(2.2)\times 10^{-5}$, which implies $\mathrm{Br}(η_1(1855)\to ηη')\sim 4.3\%$.

hep-lat↗
arXiv2023

The extreme values of two probability functions for the Gamma distribution

Motivated by Chvátal's conjecture and Tomaszewaki's conjecture, we investigate the extreme value problem of two probability functions for the Gamma distribution. Let $α,β$ be arbitrary positive real numbers and $X_{α,β}$ be a Gamma random variable with shape parameter $α$ and scale parameter $β$. We study the extreme values of functions $P\{X_{α,β}\le E[X_{α,β}]\}$ and $P\{|X_{α,β}-E[X_{α,β}]|\le \sqrt{{\rm Var}(X_{α,β})}\}$. Among other things, we show that $ \inf_{α,β}P\{X_{α,β}\le E[X_{α,β}]\}=\frac{1}{2}$ and $\inf_{α,β}P\{|X_{α,β}-E[X_{α,β}]|\le \sqrt{{\rm Var}(X_{α,β})}\}=P\{|Z|\le 1\}\approx 0.6826$, where $Z$ is a standard normal random variable.

math.PR↗
arXiv2023

A Perceptual Quality Assessment Exploration for AIGC Images

\underline{AI} \underline{G}enerated \underline{C}ontent (\textbf{AIGC}) has gained widespread attention with the increasing efficiency of deep learning in content creation. AIGC, created with the assistance of artificial intelligence technology, includes various forms of content, among which the AI-generated images (AGIs) have brought significant impact to society and have been applied to various fields such as entertainment, education, social media, etc. However, due to hardware limitations and technical proficiency, the quality of AIGC images (AGIs) varies, necessitating refinement and filtering before practical use. Consequently, there is an urgent need for developing objective models to assess the quality of AGIs. Unfortunately, no research has been carried out to investigate the perceptual quality assessment for AGIs specifically. Therefore, in this paper, we first discuss the major evaluation aspects such as technical issues, AI artifacts, unnaturalness, discrepancy, and aesthetics for AGI quality assessment. Then we present the first perceptual AGI quality assessment database, AGIQA-1K, which consists of 1,080 AGIs generated from diffusion models. A well-organized subjective experiment is followed to collect the quality labels of the AGIs. Finally, we conduct a benchmark experiment to evaluate the performance of current image quality assessment (IQA) models.

cs.CV↗
arXiv2023

Measurement of hyperfine structure and the Zemach radius in $\rm^6Li^+$ using optical Ramsey technique

We investigate the $2\,^3\!S_1$--$2\,^3\!P_J$ ($J = 0, 1, 2$) transitions in $\rm^6Li^+$ using the optical Ramsey technique and achieve the most precise values of the hyperfine splittings of the $2\,^3\!S_1$ and $2\,^3\!P_J$ states, with smallest uncertainty of about 10~kHz. The present results reduce the uncertainties of previous experiments by a factor of 5 for the $2\,^3\!S_1$ state and a factor of 50 for the $2\,^3\!P_J$ states, and are in better agreement with theoretical values. Combining our measured hyperfine intervals of the $2\,^3\!S_1$ state with the latest quantum electrodynamic (QED) calculations, the improved Zemach radius of the $\rm^6Li$ nucleus is determined to be 2.44(2)~fm, with the uncertainty entirely due to the uncalculated QED effects of order $mα^7$. The result is in sharp disagreement with the value 3.71(16) fm determined from simple models of the nuclear charge and magnetization distribution. We call for a more definitive nuclear physics value of the $\rm^6Li$ Zemach radius.

physics.atom-ph↗