arXiv subjects

Ruopeng Zhang

Publications and source records attributed to Ruopeng Zhang.

11 recordsLinked to original sources

arXiv2026

ACEsplat: Accelerated 3D Gaussian Scene Regression via RGB and Poses Only

Per-scene 3D Gaussian Splatting (3DGS) enables high-fidelity rendering, but practical robotic and AR scene capture pipelines often depend on external geometric initialization (e.g., SfM point clouds or depth estimates), which can be slow and brittle in on-site deployment. We present ACEsplat, a fast per-scene optimization framework that reconstructs 3D Gaussian representations from RGB images and camera poses only, without requiring external 3D priors (e.g., precomputed SfM models or supervised depth maps). ACEsplat uses a two-stage pipeline: (1) a self-supervised scene coordinate regression (SCR) module builds an internal geometry prior within 4--5 minutes; (2) SCR features and coordinate priors are fused by a lightweight Gaussian initialization head, followed by per-scene 3DGS optimization. On static-view rendering, ACEsplat achieves 29.11 dB PSNR on Wayspots with real-time SLAM poses and 33.20 dB on Cambridge Landmarks with SfM-refined poses. On RealEstate10K sparse-view novel view synthesis, it achieves competitive image fidelity under a challenging 2-view setting. ACEsplat completes scene-specific SCR mapping and 3DGS reconstruction within 15--25 minutes on a single GPU, making it a practical RGB+pose-only solution for rapid scene setup in robotics and mixed-reality applications.

cs.RO↗

arXiv2025

MACE: Mixture-of-Experts Accelerated Coordinate Encoding for Large-Scale Scene Localization and Rendering

Efficient localization and high-quality rendering in large-scale scenes remain a significant challenge due to the computational cost involved. While Scene Coordinate Regression (SCR) methods perform well in small-scale localization, they are limited by the capacity of a single network when extended to large-scale scenes. To address these challenges, we propose the Mixed Expert-based Accelerated Coordinate Encoding method (MACE), which enables efficient localization and high-quality rendering in large-scale scenes. Inspired by the remarkable capabilities of MOE in large model domains, we introduce a gating network to implicitly classify and select sub-networks, ensuring that only a single sub-network is activated during each inference. Furtheremore, we present Auxiliary-Loss-Free Load Balancing(ALF-LB) strategy to enhance the localization accuracy on large-scale scene. Our framework provides a significant reduction in costs while maintaining higher precision, offering an efficient solution for large-scale scene applications. Additional experiments on the Cambridge test set demonstrate that our method achieves high-quality rendering results with merely 10 minutes of training.

cs.CV↗

arXiv2024

Gravitational Wave Probe of Gravitational Dark Matter from Preheating

We forecast high-frequency gravitational wave (GW) from preheating hosting gravitational dark matter (GDM) as the indirect probe of such GDM. We use proper lattice simulations to handle resonance, and to solve GW equation of motion with the resonance induced scalar field excitations as source term. Our numerical results show that Higgs scalar excitations in Higgs preheating model give rise to magnitudes of GW energy density spectra of order $10^{-10}$ at frequencies $10-10^{3}$ MHz depending on the GDM mass, whereas inflaton fluctuation excitations in inflaton self-resonant preheating model yield magnitudes of GW energy density spectrum up to $10^{-9}~(10^{-11})$ at frequencies near $30~(2)$ MHz for the index $n=4~(6)$ with respect to the GDM mass of $1.04~(2.66)\times 10^{14}$ GeV.

hep-ph↗

arXiv2023

Gravitational freeze-in dark matter from Higgs Preheating

Gravitational freeze-in is a mechanism to explain the observed dark matter relic density if dark matter neither couples to inflation nor to standard model sector. In this work, we study gravitational freeze-in dark matter production during Higgs preheating based on non-perturbative resonance. Using reliable lattice method to handle this non-perturbative process, we show that tachyonic resonance is prohibited by strong back reaction due to Higgs self interaction needed to keep the positivity of potential during preheating, and parameter resonance is viable by tuning the Higgs self-interaction coupling to be small enough in ultraviolet energy scale. We then derive the dark matter relic density under the context of Higgs preheating, and uncover a new dark matter parameter space with dark matter mass larger than inflaton mass, which arises from out-of-equilium Higgs annihilation. Finally, we briefly remark the open question of testing gravitational dark matter.

hep-ph↗

arXiv2023

Gravitational Dark Matter from Minimal Preheating

Following our previous work, we continue to explore gravitational dark matter production during the minimal preheating caused by inflaton self-resonance. In this situation there is only one dimensionless index parameter $n$ characterizing the inflation potential after the end of inflation, which leads to a robust prediction on the gravitational dark matter relic abundance. Using lattice method to handle the non-perturbative evolutions of relevant quantities during the inflaton self-resonance, we derive the gravitational dark matter relic abundance arising from both the inflaton condensate and fluctuation annihilation. While being absent for $n=2$, the former one can instead dominate over the later one for $n=4,6$. Our results show that gravitational dark matter mass of $1.04~(2.66)\times 10^{14}$ GeV accommodates the observed value of dark matter relic abundance for $n=4$ (6).

hep-ph↗

arXiv2023

Freeze-in Dark Matter via Lepton Portal: Hubble Tension and Stellar Cooling

We propose a new freeze-in dark matter candidate which feebly couples to the standard model charged leptons. The feeble interactions allow it (i) to freeze-in from the Standard Model thermal bath with its relic density being either a fraction or the entirety of the observed dark matter density and (ii) to radiatively decay to two photons in the dark matter mass ranges of order keV scale with lifetime larger than the age of Universe. These features make this model a realistic realization of dark matter with late-time decay to reduce Hubble tension. We show the best-fit value of H_{0}=68.31(69.34) km s^{-1}Mpc^{-1} in light of Planck 2018+BAO(+LSS)+Pantheon data sets. We then use stellar cooling data to place constraints on the parameter space favored by the Hubble tension. While the universal coupling scenario is excluded, the hierarchical coupling scenario can be tested by future observations of white dwarfs after a careful look into photon inverse decay, Primakoff and Bremsstrahlung emission of the dark matter in various stellar systems. The viable parameter space may be linked to anomalies in future X-ray telescopes.

hep-ph↗

arXiv2023

Fabrication of Specimens for Atom Probe Tomography Using a Combined Gallium and Neon Focused Ion Beam Milling Approach

We demonstrate a new focused ion beam sample preparation method for atom probe tomography. The key aspect of the new method is that we use a neon ion beam for the final tip-shaping after conventional annulus milling using gallium ions. This dual-ion approach combines the benefits of the faster milling capability of the higher current gallium ion beam with the chemically inert and higher precision milling capability of the noble gas neon ion beam. Using a titanium-aluminum alloy and a layered aluminum/aluminum oxide material as test cases, we show that atom probe tips prepared using the combined gallium and neon ion approach are free from the gallium contamination that typically frustrates composition analysis of these materials due to implantation, diffusion, and embrittlement effects. We propose that by using a focused ion beam from a noble gas species, such as the neon ions demonstrated here, atom probe tomography can be more reliably performed on a larger range of materials than is currently possible using conventional techniques.

cond-mat.mtrl-sci↗

arXiv2022

Exceptional fracture toughness of CrCoNi-based medium- and high-entropy alloys close to liquid helium temperatures

Medium- and high-entropy alloys based on the CrCoNi-system have been shown to display outstanding strength, tensile ductility and fracture toughness (damage-tolerance properties), especially at cryogenic temperatures. Here we examine the JIc and (back-calculated) KJIc fracture toughness values of the face-centered cubic, equiatomic CrCoNi and CrMnFeCoNi alloys at 20 K. At flow stress values of ~1.5 GPa, crack-initiation KJIc toughnesses were found to be exceptionally high, respectively 235 and 415 MPa(square-root)m for CrMnFeCoNi and CrCoNi, with the latter displaying a crack-growth toughness Kss exceeding 540 MPa(square-root)m after 2.25 mm of stable cracking, which to our knowledge is the highest such value ever reported. Characterization of the crack-tip regions in CrCoNi by scanning electron and transmission electron microscopy reveal deformation structures at 20 K that are quite distinct from those at higher temperatures and involve heterogeneous nucleation, but restricted growth, of stacking faults and fine nano-twins, together with transformation to the hexagonal closed-packed phase. The coherent interfaces of these features can promote both the arrest and transmission of dislocations to generate respectively strength and ductility which strongly contributes to sustained strain hardening. Indeed, we believe that these nominally single-phase, concentrated solid-solution alloys develop their fracture resistance through a progressive synergy of deformation mechanisms, including dislocation glide, stacking-fault formation, nano-twinning and eventually in situ phase transformation, all of which serve to extend continuous strain hardening which simultaneously elevates strength and ductility (by delaying plastic instability), leading to truly exceptional resistance to fracture.

cond-mat.mtrl-sci↗

arXiv2022

Complex Field Inflation

We report first study of complex field inflation. Although understood as a specific two-field inflation, a complex field inflation is able to make more robust model predictions on primordial curvature perturbation. Explicitly we discuss the model realizations of complex chaotic and exponential inflation in various large-field contexts. Both complex field models contain a single complex scalar together with only two free parameters. Using numerical handles aimed to calculate primordial curvature perturbation from multifield inflation, we show that both models are compatible with current Planck data, and the individual surviving parameter space can be substantially or fully probed by future CMB-S4 experiments.

astro-ph.CO↗

arXiv2021

Twin-Boundary Structural Phase Transitions in Elemental Titanium

Twinning in crystalline materials plays an important role in many transformation and deformation processes, where underlying mechanisms can be strongly influenced by the structural, energetic and kinetic properties of associated twin boundaries (TBs). While these properties are well characterized in common cases, the possibility that TBs can display multiple complexions with distinct properties, and phase transitions between them, has not been widely explored, even though such phenomena are established in a few more general grain boundaries. We report experimental findings that {11-24} TBs in titanium display a thick interfacial region with crystalline structure distinct from the bulk. First-principles calculations establish that this complexion is linked to a metastable polymorph of titanium, and exhibits behavior consistent with a solid-state wetting transition with compressive strain, and a first-order structural transition under tension. The findings document rich TB complexion behavior in an elemental metal, with important implications for mechanical behavior and phase-transformation pathways.

cond-mat.mtrl-sci↗

arXiv2019

Verification of Short-Range Order and Its Impact on the Properties of the CrCoNi Medium Entropy Alloy

Traditional metallic alloys are mixtures of elements where the atoms of minority species tend to distribute randomly if they are below their solubility limit, or lead to the formation of secondary phases if they are above it. Recently, the concept of medium/high entropy alloys (MEA/HEA) has expanded this view, as these materials are single-phase solid solutions of generally equiatomic mixtures of metallic elements that have been shown to display enhanced mechanical properties. However, the question has remained as to how random these solid solutions actually are, with the influence of chemical short-range order (SRO) suggested in computational simulations but not seen experimentally. Here we report the first direct observation of SRO in the CrCoNi MEA using high resolution and energy-filtered transmission electron microscopy. Increasing amounts of SRO give rise to both higher stacking fault energy and hardness. These discoveries suggest that the degree of chemical ordering at the nanometer scale can be tailored through thermomechanical processing, providing a new avenue for tuning the mechanical properties of MEA/HEAs.

cond-mat.mtrl-sci↗