Spherical Harmonics Research Articles

Leveraging SO(3)-steerable convolutions for pose-robust semantic segmentation in 3D medical data

Convolutional neural networks (CNNs) allow for parameter sharing and translational equivariance by using convolutional kernels in their linear layers. By restricting these kernels to be SO(3)-steerable, CNNs can further improve parameter sharing and equivariance. These equivariant convolutional layers have several advantages over standard convolutional layers, including increased robustness to unseen poses, smaller network size, and improved sample efficiency. Despite this, most segmentation networks used in medical image analysis continue to rely on standard convolutional kernels. In this paper, we present a new family of segmentation networks that use equivariant voxel convolutions based on spherical harmonics. These SE(3)-equivariant volumetric segmentation networks, which are robust to data poses not seen during training, do not require rotation-based data augmentation during training. In addition, we demonstrate improved segmentation performance in MRI brain tumor and healthy brain structure segmentation tasks, with enhanced robustness to reduced amounts of training data and improved parameter efficiency. Code to reproduce our results, and to implement the equivariant segmentation networks for other tasks is available at <a href='http://github.com/SCAN-NRAD/e3nn_Unet'>http://github.com/SCAN-NRAD/e3nn_Unet</a>.

The Weyl expansion for the scalar and vector spherical wave functions

The Weyl expansion technique, also known as the angular spectrum expansion, expresses an outgoing spherical wave as a linear combination of plane waves. The scalar spherical waves are the solutions of the homogeneous Helmholtz equation and therefore have direct relation to the scalar multipole fields. This paper gives the Weyl expansion of multipole fields, scalar and vector, of any degree and order for spherical wave functions. The expressions are given in closed form for the scalar, , and vector, , , multipole fields, evaluated across a plane orthogonal to any given direction. In the case of scalar spherical multipoles, the spherical gradient operator has been used, while for the vector spherical multipoles, the vector spherical wave operator has been constructed.

VSHPIC: a particle-in-cell algorithm based on vector spherical harmonics expansion

The particle-in-cell (PIC) method has been widely used for studying plasma physics. However, fully three-dimensional PIC simulations always require huge computational resources. For problems with near azimuthal symmetry, recent work (Lifschitz et al 2009 J. Comput. Phys. 228 1803–14, Davidson et al 2015 J. Comput. Phys. 281 1063–77, Li et al 2021 Comput. Phys. Commun. 261 107784, Li et al 2022 J. Comput. Phys. 470 111599) has shown that expanding all the quantities defined on the grid in azimuthal harmonics and truncating the expansion can improve the code efficiency. In this paper, we describe a novel parallel algorithm for efficiently simulating three-dimensional near-spherical symmetry problems. Our approach expands all physical quantities in the θ and ϕ directions in spherical coordinates using vector spherical harmonics. The code is capable of simulating three-dimensional asymmetric scenarios by accurately tracking the evolution of distinct individual modes while preserving the charge conservation law. The fundamental dispersion relation of EM waves in the plasma has been obtained using VSHPIC simulation results. The code also shows a well strong scalability up to more than 1000 cores.

Exact massless spinor quasibound states of Schwarzschild black hole

In this work, we investigate the behavior of massless spin 12 field by working out the Dirac equation in a curved static spherically symmetric Schwarzschild space-time. Detailed derivation of novel exact massive and massless scalar quasibound state in Schwarzschild black hole background is presented in this work. After decoupling the Dirac equation, we successfully solve both of the angular and radial parts, respectively in terms of spin weighted Spherical Harmonics and Confluent Heun functions. With the exact radial wave solution in hand, applying its polynomial condition leads to the discovery of the quantized energy levels expression. We found that the massless spinor field around the Schwarzschild black hole has complex valued energy levels, in contrast to the purely imaginary for a massless boson around the same black hole. In the last section, the Hawking radiation distribution function is derived via Damour-Ruffini method and the Hawking temperature is obtained.

ZH3: Quadratic Zonal Harmonics

Spherical Harmonics (SH) have been used widely to represent lighting in games and film. While the quadratic (SH3) and higher order spherical harmonics represent irradiance well, they are expensive to store and evaluate, requiring 27 coefficients per sample. Linear SH (SH2), requiring only 12 coefficients, are sometimes used, but they do not represent irradiance signals accurately and can have challenges with negative reconstruction. We introduce a new representation (ZH3) that augments linear SH with just the zonal coefficient of quadratic SH, yielding significant visual improvement with just 15 coefficients, and discuss how solving for a luminance zonal axis can significantly improve reconstruction accuracy and reduce color artifacts. We also discuss how, rather than storing the ZH3 coefficients explicitly, we can hallucinate them from the linear SH, improving reconstruction accuracy over linear SH at minimal extra cost.

Reducing the Memory Footprint of 3D Gaussian Splatting

3D Gaussian splatting provides excellent visual quality for novel view synthesis, with fast training and realtime rendering; unfortunately, the memory requirements of this method for storing and transmission are unreasonably high. We first analyze the reasons for this, identifying three main areas where storage can be reduced: the number of 3D Gaussian primitives used to represent a scene, the number of coefficients for the spherical harmonics used to represent directional radiance, and the precision required to store Gaussian primitive attributes. We present a solution to each of these issues. First, we propose an efficient, resolution-aware primitive pruning approach, reducing the primitive count by half. Second, we introduce an adaptive adjustment method to choose the number of coefficients used to represent directional radiance for each Gaussian primitive, and finally a codebook-based quantization method, together with a half-float representation for further memory reduction. Taken together, these three components result in a x27 reduction in overall size on disk on the standard datasets we tested, along with a x1.7 speedup in rendering speed. We demonstrate our method on standard datasets and show how our solution results in significantly reduced download times when using the method on a mobile device (see Fig. 1).

Data-driven modeling of the aerodynamic deformation and drag for a freely moving drop in the sub-critical Weber number regime

Accurate prediction of the dynamics and deformation of freely moving drops is crucial for numerous droplet applications. When the Weber number is finite but below a critical value, the drop deviates from its spherical shape and deforms as it is accelerated by the gas stream. Since aerodynamic drag on the drop depends on its shape oscillation, accurately modeling the drop shape evolution is essential for predicting the drop’s velocity and position. In this study, 2D axisymmetric interface-resolved simulations were performed to provide a comprehensive dataset for developing a data-driven model. Parametric simulations were conducted by systematically varying the drop diameter and free-stream velocity, achieving wide ranges of Weber and Reynolds numbers. The instantaneous drop shapes obtained in simulations are characterized by spherical harmonics. Temporal data of the drag and modal coefficients are collected from the simulation data to train a Nonlinear Auto-Regressive models with eXogenous inputs (NARX) neural network model. The overall model consists of two multi-layer perceptron networks, which predict the modal coefficients and the drop drag, respectively. The drop shape can be reconstructed with the predicted modal coefficients. The model predictions are validated against the simulation data in the testing set, showing excellent agreement for the evolutions of both the drop shape and drag.

Perturbations of spinning black holes in dynamical Chern-Simons gravity: Slow rotation equations

The detection of gravitational waves resulting from the coalescence of binary black holes by the LIGO-Virgo-Kagra Collaboration has inaugurated a new era in gravitational physics. These gravitational waves provide a unique opportunity to test Einstein’s general relativity and its modifications in the regime of extreme gravity. A significant aspect of such tests involves the study of the ringdown phase of gravitational waves from binary black hole coalescence, which can be decomposed into a superposition of various quasinormal modes. In general relativity, the spectra of quasinormal modes depend on the mass, spin, and charge of the final black hole, but they can also be influenced by additional properties of the black hole spacetime, as well as corrections to the general theory of relativity. In this work, we focus on a specific modified theory known as dynamical Chern-Simons gravity. We employ the modified Teukolsky formalism developed in a previous study and lay down the foundations to investigate perturbations of slowly rotating black holes admitted by the theory. Specifically, we derive the master equations for the Ψ0 and Ψ4 Weyl scalar perturbations that characterize the radiative part of gravitational perturbations, as well as the master equation for the scalar field perturbations. We employ metric reconstruction techniques to obtain explicit expressions for all relevant quantities. Finally, by leveraging the properties of spin-weighted spheroidal harmonics to eliminate the angular dependence from the evolution equations, we derive two, radial, second-order, ordinary differential equations for Ψ0 and Ψ4, respectively. These two equations are coupled to another radial, second-order, ordinary differential equation for the scalar field perturbations. This work is the first attempt to derive a master equation for black holes in dynamical Chern-Simons gravity using curvature perturbations. The master equations we obtain can then be numerically integrated to obtain the quasinormal mode spectrum of slowly rotating black holes in this theory, making progress in the study of ringdown in dynamical Chern-Simons gravity. Published by the American Physical Society 2024

Relativistic massive and massless scalar fields bound to the Bumblebee gravity's black hole

In this letter, we examine massive and massless scalar quasibound states bound in the static spherically symmetric black hole solution of the Einstein-Bumblebee space-time. We start with constructing the governing relativistic fields' dynamical equation, i.e. the Klein-Gordon equation, component by component. With the help of the ansatz of separation of variables, we find the exact solution of the angular part in terms of Spherical Harmonics while the radial exact solution is discovered in terms of the Confluent Heun function. The quantization of the quasibound state is done by applying the polynomial condition of the Confluent Heun function that reveals a complex-valued energy levels expression for the case of massive scalar, where the real part is the scalar particle's energy and the imaginary part represents the quasibound state's decay. And for a massless scalar, a pure imaginary energy levels is obtained. The quasibound states, thus, describe decaying relativistic states bound in the black hole's gravitational potential well. At the end, we investigate the Hawking radiation of the static Bumblebee black hole's horizon by deriving and investigating the radiation distribution function.

Reconstruction of the initial data from the trace of the solutions on an infinite time cylinder of damped wave equations *

In this paper, we consider two types of damped wave equations: the weakly damped equation and the strongly damped equation. We recover the initial velocity from the trace of the solution on a space-time cylinder. This inverse problem is related to photoacoustic tomography (PAT), a hybrid medical imaging technique. PAT is based on generating acoustic waves inside of an object of interest and one of the mathematical problem in PAT is reconstructing the initial velocity from the solution of the wave equation measured on the outside of object. Using the spherical harmonics and spectral theorem, we demonstrate a way to recover the initial velocity.

Open‐source stand‐alone version of atmospheric model Aeolus 2.0 software

AbstractIn this discourse, we present the unveiling of an open‐source software package designed to facilitate engagement with the atmospheric model, Aeolus 2.0. This particular iteration stands as a self‐contained model of intermediate complexity. The model's dynamical core is underpinned by a multi‐layer pseudo‐spectral moist‐convective Thermal Rotating Shallow Water (mcTRSW) model. The pseudo‐spectral problem‐solving tasks are handled by the Dedalus algorithm, acknowledged for its spin‐weighted spherical harmonics. The model captures the temporal and spatial evolution of vertically integrated potential temperature, thickness, water vapour, precipitation, and the intricate influence of bottom topography. It comprehensively characterizes velocity fields in both the lower and upper troposphere, employing resolutions spanning a spectrum from the smooth to the coarse, enabling the exploration of a wide range of dynamic phenomena with varying levels of detail and precision.

Differentiable design of a double-freeform lens with multi-level radial basis functions for extended source irradiance tailoring

Freeform optics are key for generating prescribed illumination patterns from given sources, which are crucial for solid-state lighting and machine vision illumination. There is an increasing demand for compact freeform optics, which presents a substantial challenge for current design methods since the source dimensions must be considered. Most current extended-source design methods, although requiring profound knowledge of optics and mathematics, focus on the modest goal of obtaining uniform irradiance distributions. We address a more challenging design problem of generating an irradiance distribution of arbitrary shape through a double-freeform lens that can fully encompass the extended source. We propose a differentiable design method whose uniqueness lies in the representation of the double-freeform surfaces using multi-level spherical radial basis functions, which has a natural link to a multi-scale optimization technique. In addition, we employ a sequential unconstrained minimization technology complemented with Lagrange multipliers that add key feasibility constraints on lens shape and size. The proposed method is flexible, general, and efficient in designing highly compact freeform lenses for generating both simple and complex irradiance distributions, as demonstrated through the design examples. This could enable a universal solution to the extended-source design problem.

SDUST2020MGCR: a global marine gravity change rate model determined from multi-satellite altimeter data

Abstract. Investigating the global time-varying gravity field mainly depends on GRACE/GRACE-FO gravity data. However, satellite gravity data exhibit low spatial resolution and signal distortion. Satellite altimetry is an important technique for observing the global ocean and provides many consecutive years of data, which enables the study of high-resolution marine gravity variations. This study aims to construct a high-resolution marine gravity change rate (MGCR) model using multi-satellite altimetry data. Initially, multi-satellite altimetry data and ocean temperature–salinity data from 1993 to 2019 are utilized to estimate the altimetry sea level change rate (SLCR) and steric SLCR, respectively. Subsequently, the mass-term SLCR is calculated. Finally, based on the mass-term SLCR, the global MGCR model on 5′ × 5′ grids (SDUST2020MGCR) is constructed by applying the spherical harmonic function method and mass load theory. Comparisons and analyses are conducted between SDUST2020MGCR and GRACE2020MGCR resolved from GRACE/GRACE-FO gravity data. The spatial distribution characteristics of SDUST2020MGCR and GRACE2020MGCR are similar in the sea areas where gravity changes significantly, such as the eastern seas of Japan, the western seas of the Nicobar Islands, and the southern seas of Greenland. The statistical mean values of SDUST2020MGCR and GRACE2020MGCR in global and local oceans are all positive, indicating that MGCR is rising. Nonetheless, differences in spatial distribution and statistical results exist between SDUST2020MGCR and GRACE2020MGCR, primarily attributable to spatial resolution disparities among altimetry data, ocean temperature–salinity data, and GRACE/GRACE-FO data. Compared with GRACE2020MGCR, SDUST2020MGCR has higher spatial resolution and excludes stripe noise and leakage errors. The high-resolution MGCR model constructed using altimetry data can reflect the long-term marine gravity change in more detail, which is helpful in studying seawater mass migration and its associated geophysical processes. The SDUST2020MGCR model data are available at https://doi.org/10.5281/zenodo.10701641 (Zhu et al., 2024).

Clustering of dark matter in the cosmic web as a probe of massive neutrinos

ABSTRACT The large-scale structure of the Universe is distributed in a cosmic web. Studying the distribution and clustering of dark matter particles and haloes may open up a new horizon for studying the physics of the dark Universe. In this work, we investigate the nearest neighbour statistics and spherical contact function in cosmological models with massive neutrinos. For this task, we use the relativistic N-body code, gevolution, and study particle snapshots at three different redshifts. In each snapshot, we find the haloes and evaluate the letter functions for them. We show that a generic behaviour can be found in the nearest neighbour, G(r), and spherical contact functions, F(r), which makes these statistics promising tools to constrain the total neutrino mass.

A G-Modified Helmholtz Equation with New Expansions for the Earth’s Disturbing Gravitational Potential, Its Functionals and the Study of Isogravitational Surfaces

The G-modified Helmholtz equation is a partial differential equation that enables us to express gravity intensity g as a series of spherical harmonics having radial distance r in irrational powers. The Laplace equation in three-dimensional space (in Cartesian coordinates, is the sum of the second-order partial derivatives of the unknown quantity equal to zero) is used to express the Earth’s gravity potential (disturbing and normal potential) in order to represent other useful quantities—which are also known as functionals of the disturbing potential—such as gravity disturbance, gravity anomaly, and geoid undulation as a series of spherical harmonics. We demonstrate that by using the G-modified Helmholtz equation, not only gravity intensity but also disturbing potential and its functionals can be expressed as a series of spherical harmonics. Having gravity intensity represented as a series of spherical harmonics allows us to create new Global Gravity Models. Furthermore, a more detailed examination of the Earth’s isogravitational surfaces is conducted. Finally, we tabulate our results, which makes it clear that new Global Gravity Models for gravity intensity g will be very useful for many geophysical and geodetic applications.

Efficient computation of the geopotential gradient in graphic processing units

Efficient computation of the geopotential gradient is essential for numerical propagators, particularly in scenarios involving low Earth orbits. Conventional geopotential calculations are based on spherical harmonics series, which become computationally demanding as the degree/order increases. This computational burden can be mitigated by means of parallelized algorithms. Additionally, certain situations lend themselves to high parallelization, such as the propagation of space debris catalogs, satellite mega-constellations, or the dispersion of particles resulting from a space collision event. This paper introduces an optimized Graphics Processing Unit (GPU) implementation designed to facilitate extensive parallelization in the geopotential gradient calculation. The formulation developed in this study is not specific to any GPU. However, to illustrate the low-level optimizations necessary for an efficient implementation, we selected the Compute Unified Device Architecture (CUDA) as the dominant and de facto standard in parallel computing. Nevertheless, most of the concepts and optimizations presented in this paper are also valid for other GPU architectures. Built upon the spherical harmonic expansion using the Cunningham formulation, which is well-suited for GPU computations, our implementation offers several variants with different tradeoffs between speed and accuracy. Besides GPU double precision, we introduced a mixed precision arithmetic –a hybrid between single and double precision– that exploits GPU capabilities with a low penalty in accuracy. The proposed algorithm was implemented as a software reusable module, and its performance was evaluated against GMAT, GODOT, and Orekit astrodynamic codes. The algorithm’s accuracy in double precision is comparable to such codes. The mixed precision version showed enough accuracy for LEO satellite propagation, with around 1 m difference in four days. Testing across different CUDA architectures revealed very high speed-up factors compared to a single CPU, reaching a speed-up of 645 for the mixed precision variant and 450 for the double precision one in the propagation of about 3200 objects with a geopotential of degree/order 126 × 126 using an A100 GPU device.

A 3-D minimum-enstrophy vortex in stratified quasi-geostrophic flows

Applying a variational analysis, a minimum-enstrophy vortex in three-dimensional (3-D) fluids with continuous stratification is found, under the quasi-geostrophic hypothesis. The buoyancy frequency is held constant. This vortex is an ideal limiting state in a flow with an enstrophy decay while energy and generalized angular momentum remain fixed. The variational method used to obtain two-dimensional (2-D) minimum-enstrophy vortices is applied here to 3-D integral quantities. The solution from the first-order variation is expanded on a basis of orthogonal spherical Bessel functions. By computing second-order variations, the solution is found to be a true minimum in enstrophy. This solution is weakly unstable when inserted in a numerical code of the quasi-geostrophic equations. After a stage of linear instability, nonlinear wave interaction leads to the reorganization of this vortex into a tripolar vortex. Further work will relate our solution with maximal entropy 3-D vortices.

Fourier-Hankel-Abel Nyquist-limited tomography: A spherical harmonic basis function approach to tomographic velocity-map image reconstruction.

A new method for the fully generalized reconstruction of three-dimensional (3D) photoproduct distributions from velocity-map imaging (VMI) projection data is presented. This approach, dubbed Fourier-Hankel-Abel Nyquist-limited TOMography (FHANTOM), builds on recent previous work in tomographic image reconstruction [C. Sparling and D. Townsend, J. Chem. Phys. 157, 114201 (2022)] and takes advantage of the fact that the distributions produced in typical VMI experiments can be simply described as a sum over a small number of spherical harmonic functions. Knowing the solution is constrained in this way dramatically simplifies the reconstruction process and leads to a considerable reduction in the number of projections required for robust tomographic analysis. Our new method significantly extends basis set expansion approaches previously developed for the reconstruction of photoproduct distributions possessing an axis of cylindrical symmetry. FHANTOM, however, can be applied generally to any distribution-cylindrically symmetric or otherwise-that can be suitably described by an expansion in spherical harmonics. Using both simulated and real experimental data, this new approach is tested and benchmarked against other tomographic reconstruction strategies. In particular, the reconstruction of photoelectron angular distributions recorded in a strong-field ionization regime-marked by their extensive expansion in terms of spherical harmonics-serves as a key test of the FHANTOM methodology. With the increasing use of exotic optical polarization geometries in photoionization experiments, it is anticipated that FHANTOM and related reconstruction techniques will provide an easily accessible and relatively low-cost alternative to more advanced 3D-VMI spectrometers.

Limits of Bessel functions for root systems as the rank tends to infinity

We study the asymptotic behaviour of Bessel functions associated to root systems of type An−1 and type Bn with positive multiplicities as the rank n tends to infinity. In both cases, we characterize the possible limit functions and the Vershik–Kerov type sequences of spectral parameters for which such limits exist. In the type A case, this gives a new and very natural approach to recent results by Assiotis and Najnudel in the context of β-ensembles in random matrix theory. These results generalize known facts about the approximation of the positive-definite Olshanski spherical functions of the space of infinite-dimensional Hermitian matrices over F=R,ℂ,H (with the action of the associated infinite unitary group) by spherical functions of finite-dimensional spaces of Hermitian matrices. In the type B case, our results include asymptotic results for the spherical functions associated with the Cartan motion groups of non-compact Grassmannians as the rank goes to infinity, and a classification of the Olshanski spherical functions of the associated inductive limits.

Near-Field Source Localization and Beamforming in the Spherical Sector Harmonics Domain

Near-Field Source Localization and Beamforming in the Spherical Sector Harmonics Domain