arXiv subjects

Jun Jiang

Publications and source records attributed to Jun Jiang.

At least 91 records · Page 5Linked to original sources

arXiv2023

Deformations of modified $r$-matrices and cohomologies of related algebraic structures

Modified $r$-matrices are solutions of the modified classical Yang-Baxter equation, introduced by Semenov-Tian-Shansky, and play important roles in mathematical physics. In this paper, first we introduce a cohomology theory for modified $r$-matrices. Then we study three kinds of deformations of modified $r$-matrices using the established cohomology theory, including algebraic deformations, geometric deformations and linear deformations. We give the differential graded Lie algebra that governs algebraic deformations of modified $r$-matrices. For geometric deformations, we prove the rigidity theorem and study when is a neighborhood of a modified $r$-matrix smooth in the space of all modified $r$-matrix structures. In the study of trivial linear deformations, we introduce the notion of a Nijenhuis element for a modified $r$-matrix. Finally, applications are given to study deformations of complement of the diagonal Lie algebra and compatible Poisson structures.

math-ph↗

arXiv2023

Category Feature Transformer for Semantic Segmentation

Aggregation of multi-stage features has been revealed to play a significant role in semantic segmentation. Unlike previous methods employing point-wise summation or concatenation for feature aggregation, this study proposes the Category Feature Transformer (CFT) that explores the flow of category embedding and transformation among multi-stage features through the prevalent multi-head attention mechanism. CFT learns unified feature embeddings for individual semantic categories from high-level features during each aggregation process and dynamically broadcasts them to high-resolution features. Integrating the proposed CFT into a typical feature pyramid structure exhibits superior performance over a broad range of backbone networks. We conduct extensive experiments on popular semantic segmentation benchmarks. Specifically, the proposed CFT obtains a compelling 55.1% mIoU with greatly reduced model parameters and computations on the challenging ADE20K dataset.

cs.CV↗

arXiv2023

Charm-Meson $t$-channel Singularities in an Expanding Hadron Gas

We study the time evolution of the numbers of charm mesons after the kinetic freezeout of the expanding hadron gas produced by the hadronization of the quark-gluon plasma from a central heavy-ion collision. The $πD$ reaction rates have contributions from a $D^\ast$ resonance in the $s$ channel. The $πD^\ast$ reaction rates are enhanced by $t$-channel singularities from an intermediate $D$. The contributions to reaction rates from $D^\ast$ resonances and $D$-meson $t$-channel singularities are sensitive to thermal mass shifts and thermal widths. In the expanding hadron gas, the $t$-channel singularities are regularized by the thermal $D$ widths. After kinetic freezeout, the thermal $D$ widths are dominated by coherent pion forward scattering. The contributions to $πD^\ast$ reaction rates from $t$-channel singularities are inversely proportional to the pion number density, which decreases to 0 as the hadron gas expands. The $t$-channel singularities produce small but significant changes in charm-meson ratios from those predicted using the known $D^\ast$-decay branching fractions.

hep-ph↗

arXiv2023

Single Image Deraining via Feature-based Deep Convolutional Neural Network

It is challenging to remove rain-steaks from a single rainy image because the rain steaks are spatially varying in the rainy image. Although the CNN based methods have reported promising performance recently, there are still some defects, such as data dependency and insufficient interpretation. A single image deraining algorithm based on the combination of data-driven and model-based approaches is proposed. Firstly, an improved weighted guided image filter (iWGIF) is used to extract high-frequency information and learn the rain steaks to avoid interference from other information through the input image. Then, transfering the input image and rain steaks from the image domain to the feature domain adaptively to learn useful features for high-quality image deraining. Finally, networks with attention mechanisms is used to restore high-quality images from the latent features. Experiments show that the proposed algorithm significantly outperforms state-of-the-art methods in terms of both qualitative and quantitative measures.

cs.CV↗

arXiv2023

Relations for low-energy constants in baryon chiral perturbation theory with explicit $Δ(1232)$ derived from the chiral quark model

We study the relations between low-energy constants (LECs) in the chiral Lagrangians with $Δ(1232)$ and those in the quark-level description model up to the third chiral order. Ten structure correspondences are involved in getting the relations. This situation is more complicated than the spin-1/2 baryon case. The obtained results may help to further investigations involving the $Δ(1232)$ baryons.

hep-ph↗

arXiv2023

Production of double heavy quarkonium at super $Z$ factory

Within the color singlet model, we calculate the exclusive production of double charmonia, double bottomonia, and double $B_c$ mesons at future super $Z$ factory. The two heavy quarkonia or $B_c$'s are either two S-wave Fock states ($^1S_0, ~^3S_1$), or one S-wave and one P-wave states ($^1P_1,~^3P_J~(J=0,1,2)$). The top three $Z^0$ propagated channels in cross sections for double charmonia are $J/ψ+h_c,η_c+χ_{c2}$, and $η_c+χ_{c0}$. For double bottomonia, they are $η_b+Υ, Υ+χ_{b2}$, and $Υ+Υ$. For double $B_c$ mesons, they are $B_c^{*+}+B_c^{*-}$,$B_c^{*+}+χ_{bc2}^-$, and $η_{bc}^++B_c^{*-}$. The cross sections of double $B_c$ mesons are roughly one order of magnitude larger than those of the double bottomonia, and two orders of magnitude larger than those of the double charmonia. To make it helpful for experimental study, we present the total cross sections $σ$ as functions of CM energy $\sqrt{s}$, $σ$ as functions of the renormalization scale $μ$, the angle distributions $dσ/dcosθ$, and the $p_T$ distributions $dσ/dp_{t}$. The uncertainties come from the varying masses of constituent heavy quarks bring up to 20% corrections. We also find that the initial state radiation can bring about 30%$\sim$40% suppresions when 1\%$m_Z$ energy is losing, and cross sections can increase by about $2\sim3$ times or decrease by an order of magnitude when adopting different potential models which becomes the major source of uncertainty. The numerical results show that it might be not optimistic for the experimental observation, but it is still far from excluded at the FCC-ee and also the CEPC running in the $Z$ factory mode.

hep-ph↗

arXiv2022

Be optical lattice clocks with the fractional Stark shift up to the level of 10$^{-19}$

The energy levels and electric dipole ($E1$) matrix elements of the ground state and low-lying excited states of Be atoms are calculated using the relativistic configuration interaction plus core polarization (RCICP) method. The static and dynamic $E1$, magnetic dipole ($M1$) and electric quadrupole ($E2$) polarizabilities as well as the hyperpolarizabilities of the $2s^2~^1S_0$ and $2s2p~^3P_0$ states are determined. Two magic wavelengths, 300.03 and 252.28 nm, of $2s^2~^1S_0 \rightarrow 2s2p~^3P_0$ clock transition are found. Then, the multipolar and nonlinear Stark shifts of the clock transition at the magic wavelength are discussed in detail. We find that when the laser intensity $I$ is in the range of 14.3 $\sim$ 15.9 kW/cm$^{2}$ and the detuning $δ$ (the frequency detuning of the lattice laser frequency relative to the magic frequency) is in the range of 40.7 $\sim$ 40.9 MHz, the fractional Stark shifts of the clock transition are less than 1.0 $\times$ 10$^{-18}$. While, when $I$ is in the range of 15.01 $\sim$ 15.46 kW/cm$^{2}$ and $δ$ is in the range of 40.73 $\sim$ 40.76 MHz, the fractional Stark shifts are lower than 1.0 $\times$ 10$^{-19}$.

physics.atom-ph↗

arXiv2022

MIPI 2022 Challenge on Under-Display Camera Image Restoration: Methods and Results

Developing and integrating advanced image sensors with novel algorithms in camera systems are prevalent with the increasing demand for computational photography and imaging on mobile platforms. However, the lack of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). To bridge the gap, we introduce the first MIPI challenge including five tracks focusing on novel image sensors and imaging algorithms. In this paper, we summarize and review the Under-Display Camera (UDC) Image Restoration track on MIPI 2022. In total, 167 participants were successfully registered, and 19 teams submitted results in the final testing phase. The developed solutions in this challenge achieved state-of-the-art performance on Under-Display Camera Image Restoration. A detailed description of all models developed in this challenge is provided in this paper. More details of this challenge and the link to the dataset can be found at https://github.com/mipi-challenge/MIPI2022.

eess.IV↗

arXiv2022

Cohomologies of difference Lie groups and van Est theorem

A difference Lie group is a Lie group equipped with a difference operator, equivalently a crossed homomorphism with respect to the adjoint action. In this paper, first we introduce the notion of a representation of a difference Lie group, and establish the relation between representations of difference Lie groups and representations of difference Lie algebras via differentiation and integration. Then we introduce a cohomology theory for difference Lie groups and justify it via the van Est theorem. Finally, we classify abelian extensions of difference Lie groups using the second cohomology group as applications.

math.RA↗

arXiv2022

Reversed Image Signal Processing and RAW Reconstruction. AIM 2022 Challenge Report

Cameras capture sensor RAW images and transform them into pleasant RGB images, suitable for the human eyes, using their integrated Image Signal Processor (ISP). Numerous low-level vision tasks operate in the RAW domain (e.g. image denoising, white balance) due to its linear relationship with the scene irradiance, wide-range of information at 12bits, and sensor designs. Despite this, RAW image datasets are scarce and more expensive to collect than the already large and public RGB datasets. This paper introduces the AIM 2022 Challenge on Reversed Image Signal Processing and RAW Reconstruction. We aim to recover raw sensor images from the corresponding RGBs without metadata and, by doing this, "reverse" the ISP transformation. The proposed methods and benchmark establish the state-of-the-art for this low-level vision inverse problem, and generating realistic raw sensor readings can potentially benefit other tasks such as denoising and super-resolution.

eess.IV↗

arXiv2022

Overexposure Mask Fusion: Generalizable Reverse ISP Multi-Step Refinement

With the advent of deep learning methods replacing the ISP in transforming sensor RAW readings into RGB images, numerous methodologies solidified into real-life applications. Equally potent is the task of inverting this process which will have applications in enhancing computational photography tasks that are conducted in the RAW domain, addressing lack of available RAW data while reaping from the benefits of performing tasks directly on sensor readings. This paper's proposed methodology is a state-of-the-art solution to the task of RAW reconstruction, and the multi-step refinement process integrating an overexposure mask is novel in three ways: instead of from RGB to bayer, the pipeline trains from RGB to demosaiced RAW allowing use of perceptual loss functions; the multi-step processes has greatly enhanced the performance of the baseline U-Net from start to end; the pipeline is a generalizable process of refinement that can enhance other high performance methodologies that support end-to-end learning.

cs.CV↗

arXiv2022

MIPI 2022 Challenge on RGBW Sensor Fusion: Dataset and Report

Developing and integrating advanced image sensors with novel algorithms in camera systems are prevalent with the increasing demand for computational photography and imaging on mobile platforms. However, the lack of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). To bridge the gap, we introduce the first MIPI challenge, including five tracks focusing on novel image sensors and imaging algorithms. In this paper, RGBW Joint Fusion and Denoise, one of the five tracks, working on the fusion of binning-mode RGBW to Bayer, is introduced. The participants were provided with a new dataset including 70 (training) and 15 (validation) scenes of high-quality RGBW and Bayer pairs. In addition, for each scene, RGBW of different noise levels was provided at 24dB and 42dB. All the data were captured using an RGBW sensor in both outdoor and indoor conditions. The final results are evaluated using objective metrics, including PSNR, SSIM}, LPIPS, and KLD. A detailed description of all models developed in this challenge is provided in this paper. More details of this challenge and the link to the dataset can be found at https://github.com/mipi-challenge/MIPI2022.

eess.IV↗

arXiv2022

MIPI 2022 Challenge on RGB+ToF Depth Completion: Dataset and Report

Developing and integrating advanced image sensors with novel algorithms in camera systems is prevalent with the increasing demand for computational photography and imaging on mobile platforms. However, the lack of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). To bridge the gap, we introduce the first MIPI challenge including five tracks focusing on novel image sensors and imaging algorithms. In this paper, RGB+ToF Depth Completion, one of the five tracks, working on the fusion of RGB sensor and ToF sensor (with spot illumination) is introduced. The participants were provided with a new dataset called TetrasRGBD, which contains 18k pairs of high-quality synthetic RGB+Depth training data and 2.3k pairs of testing data from mixed sources. All the data are collected in an indoor scenario. We require that the running time of all methods should be real-time on desktop GPUs. The final results are evaluated using objective metrics and Mean Opinion Score (MOS) subjectively. A detailed description of all models developed in this challenge is provided in this paper. More details of this challenge and the link to the dataset can be found at https://github.com/mipi-challenge/MIPI2022.

cs.CV↗

arXiv2022

MIPI 2022 Challenge on Quad-Bayer Re-mosaic: Dataset and Report

Developing and integrating advanced image sensors with novel algorithms in camera systems are prevalent with the increasing demand for computational photography and imaging on mobile platforms. However, the lack of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). To bridge the gap, we introduce the first MIPI challenge, including five tracks focusing on novel image sensors and imaging algorithms. In this paper, Quad Joint Remosaic and Denoise, one of the five tracks, working on the interpolation of Quad CFA to Bayer at full resolution, is introduced. The participants were provided a new dataset, including 70 (training) and 15 (validation) scenes of high-quality Quad and Bayer pairs. In addition, for each scene, Quad of different noise levels was provided at 0dB, 24dB, and 42dB. All the data were captured using a Quad sensor in both outdoor and indoor conditions. The final results are evaluated using objective metrics, including PSNR, SSIM, LPIPS, and KLD. A detailed description of all models developed in this challenge is provided in this paper. More details of this challenge and the link to the dataset can be found at https://github.com/mipi-challenge/MIPI2022.

eess.IV↗

arXiv2022

MIPI 2022 Challenge on RGBW Sensor Re-mosaic: Dataset and Report

Developing and integrating advanced image sensors with novel algorithms in camera systems are prevalent with the increasing demand for computational photography and imaging on mobile platforms. However, the lack of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). To bridge the gap, we introduce the first MIPI challenge including five tracks focusing on novel image sensors and imaging algorithms. In this paper, RGBW Joint Remosaic and Denoise, one of the five tracks, working on the interpolation of RGBW CFA to Bayer at full resolution, is introduced. The participants were provided with a new dataset including 70 (training) and 15 (validation) scenes of high-quality RGBW and Bayer pairs. In addition, for each scene, RGBW of different noise levels was provided at 0dB, 24dB, and 42dB. All the data were captured using an RGBW sensor in both outdoor and indoor conditions. The final results are evaluated using objective metrics including PSNR, SSIM, LPIPS, and KLD. A detailed description of all models developed in this challenge is provided in this paper. More details of this challenge and the link to the dataset can be found at https://github.com/mipi-challenge/MIPI2022.

cs.CV↗

arXiv2022

Relations for low-energy coupling constants in baryon chiral perturbation theory derived from the chiral quark model

The quark model symmetry can be adopted to establish relations between the low-energy constants (LECs) in the baryon chiral perturbation theory ($χPT$) if one assumes that a baryon-baryon-meson coupling is described equivalently by a quark-quark-meson coupling at the quark level. Through the correspondence between the $SU(2)$ description and the $SU(3)$ description for the same coupling vertex at the quark level, we find some relations between the LECs in $SU(2)_{χPT}$ and $SU(3)_{χPT}$ up to the third chiral order. The $SU(3)_{χPT}$ LEC relations at the same order are also obtained. The numerical analysis roughly supports these relations. In the situation that the available experimental data are not enough, one may employ such constraints to reduce the number of LECs.

hep-ph↗

arXiv2022

Evolution of charm-meson ratios in an expanding hadron gas

We study the time evolution of the numbers of charm mesons after the kinetic freeze-out of the hadron gas produced by a central heavy-ion collision. The $πD^\ast \to πD^\ast$ reaction rates have $t$-channel singularities that give contributions inversely proportional to the thermal width of the $D$. The ratio of the $D^0$ and $D^+$ production rates can differ significantly from those predicted using the measured $D^\ast$ branching fractions.

hep-ph↗

arXiv2022

Triangle Singularity in the Production of $T_{cc}^+$(3875) and a Soft Pion

The double-charm tetraquark meson $T_{cc}^+(3875)$ can be produced in high-energy proton-proton collisions by the creation of the charm mesons $D^{*+} D^0$ at short distances followed by their binding into $T_{cc}^+$. The $T_{cc}^+$ can also be produced by the creation of $D^{*+} D^{*+}$ at short distances followed by their rescattering into $T_{cc}^+ π^+$. A charm-meson triangle singularity produces a narrow peak in the $T_{cc}^+ π^+$ invariant mass distribution 6.1 MeV above the threshold with a width of about 1 MeV. Well beyond the peak, the differential cross section decreases with the invariant kinetic energy $E$ of $T_{cc}^+ π^+$ as $E^{-1/2}$. The fraction of $T_{cc}^+$ that are accompanied by $π^+$ with $E< m_π$ is estimated to be roughly 3%. The fraction of $T_{cc}^+$ events with $T_{cc}^+ π^+$ in the narrow peak from the triangle singularity could be comparable.

hep-ph↗