We propose PanoStyleSuite, a unified end-to-end framework for panoramic indoor style transfer that integrates photorealistic generation, real-world adaptability, and immersive WebXR-based exploration. Starting from a single 360° panorama, our pipeline employs a multitask dense prediction transformer to infer semantic segmentation, depth, albedo, shading, and layout—providing geometry- and photometry-aware cues for shading-independent stylization. A geometry-aware GAN then applies photorealistic style transfer constrained by depth-, layout-, and edge-consistency losses, while an enhanced intrinsic decomposition model explicitly handles both obscurance and highlight components to ensure illumination-stable stylization. Super-resolution further restores high-frequency details suitable for VR deployment. Building upon this foundation, we introduce an immersive WebXR extension that converts stylized panoramas into stereoscopic multi-center-of-projection views, enabling head-tracked VR exploration. A semantic-aware style embedding, computed from per-class albedo-driven descriptors, supports metric-based interactive in-VR recommendation and real-time style selection. Experiments on Structured3D dataset demonstrate state-of-the-art performance in perceptual fidelity (PSNR, SSIM) and stylistic realism (ArtFID), while qualitative evaluations on Stanford2D3D dataset confirm robust in-the-wild generalization enabled by multitask inference. A VR-based user study with 21 participants indicates that participants assigned higher ratings to the proposed geometry-aware variants across five perceptual dimensions; one-way ANOVA reveals statistically significant differences for Visual Quality, Semantic Awareness, Style Appeal, Seamlessness, and Overall Satisfaction (all p<10−3), underscoring the critical importance of geometry-aware constraints for perceptual realism in immersive environments. Overall, PanoStyleSuite provides the first fully integrated solution spanning photorealistic stylization, real-world adaptability, and immersive WebXR interaction—advancing applications in virtual staging, interior design, real-estate visualization, and metaverse-oriented environments.
Magnetite (Fe3O4) nanoparticles (NPs) were obtained through polyol synthesis followed by a particle-growth treatment at 260 °C, without external surfactants, to evaluate the method's efficiency for producing magnetic nanoparticles with a well-defined crystalline structure, colloidal stability, and surface functionalization. The NPs were characterized by X-ray diffraction, Fourier-transform infrared spectroscopy, thermal analyses, dynamic light scattering, transmission electron microscopy, Mössbauer spectroscopy, and magnetometry. X-ray diffraction confirmed predominant magnetite formation, with a cubic spinel-type structure and an average crystallite size of approximately 16.75 nm. The FTIR and TG/DSC analyses indicated the presence of organic groups derived from Terathane on the nanoparticles' surface, highlighting the polyol's role as a stabilizing agent. No visible sedimentation or macroscopic aggregation was observed in ethanol, THF, or chloroform after 45 days. DLS analysis indicated an average hydrodynamic diameter of 31 nm, while TEM images revealed nearly spherical particles with an average diameter of approximately 10.41 nm. Mössbauer spectroscopy confirmed the predominance of magnetite and indicated magnetic relaxation effects, while magnetometry revealed a magnetically soft response, with a saturation magnetization of 54.91 emu·g−1, a coercive field of 0.188 kOe, and a remanent magnetization of 12.63 emu·g−1. Thus, the proposed route proved to be simple and efficient for synthesizing Fe3O4 nanoparticles with promising properties for applications in magnetic colloids, nanocomposites, environmental remediation, and biomedical technologies.
The Lieb lattice, featuring the coexistence of Dirac-like dispersions and a perfectly flat band, constitutes a paradigmatic platform for investigating the interplay between band topology, many-body interactions, and external perturbations. In this work, we present a comprehensive theoretical study of the combined effects of Holstein electron–phonon coupling and a perpendicular magnetic field on the dynamical and static charge structure factors of a doped Lieb lattice. Employing a full-band Green’s function formalism combined with the random-phase approximation, we derive the phonon-mediated effective electron–electron interaction and compute the charge susceptibilities. Our results reveal that the frequency, intensity, and spectral weight of plasmon peaks in the dynamical charge structure factor can be systematically tuned. The controlling parameters are the electron–phonon coupling strength, Zeeman field, next-nearest-neighbor hopping, staggered on-site potential, and carrier doping. In particular, increasing the Holstein coupling induces a pronounced blueshift of the high-frequency plasmon mode. This blueshift is accompanied by a non-monotonic variation of the mode intensity. In contrast, the perpendicular magnetic field produces a redshift and a suppression of the plasmon resonance. The static charge structure factor exhibits rich non-monotonic dependencies on magnetic field and hopping amplitude. These dependencies reflect the competition between thermal fluctuations, spin polarization, and polaronic effects. These findings establish a unified framework for the control of collective charge excitations and charge correlations via Holstein electron–phonon coupling and perpendicular magnetic fields. The resulting tunability of plasmon frequencies, spectral weights, and static charge correlations offers concrete design principles for reconfigurable plasmonic responses and engineered charge correlations in artificial quantum lattices and photonic platforms.
Bimetallic nanowires (NWs) integrating magnetic, electrical, and catalytic functionalities are of interest for multifunctional nanostructured systems. Herein, Ag@Ni core-shell NWs were synthesized through a two-step route using preformed Ag NWs as the core and a Ni-containing shell. SEM, TEM, EDS elemental mapping, line-scan analysis, and XRD collectively supported the formation of an Ag-rich core surrounded by a spike-decorated polycrystalline Ni-containing shell. Magnetic hysteresis measurements revealed ferromagnetic behavior at 300 and 10 K, with coercivities of approximately 53 and 80 Oe, respectively. ZFC/FC measurements under 100 Oe showed thermomagnetic irreversibility, with the two curves becoming nearly coincident near the upper end of the measured temperature range of 340–350 K. Representative individual-NW devices exhibited approximately linear I-V characteristics under zero magnetic field, and the effective resistivity of the Ag@Ni NWs was 4.4×10−4Ω⋅cm, intermediate between those of Ag and Ni NWs. In NaBH4 hydrolysis, the Ag@Ni NWs showed the fastest overall hydrogen-evolution profile among the tested samples under identical total-catalyst-mass conditions. These results show that the prepared Ag@Ni core-shell NWs exhibit ferromagnetic behavior, electrical conduction, and catalytic hydrogen-generation activity under the respective measurement conditions.
As early as in 1988, the IEC discussed the application of tangential field coil (H-coil) for physically more correct Single Sheet Testers. The reason was the starting insight that the calculation of the magnetic field strength H(t) from the magnetization current i(t) by a nominal path length LM involves a systematic source of error that impedes correct loss determination, a priori. The main aim of the current paper is to analyze and compare in the literature existing H-coil types. For most accurate measurement, the H-coil should be arranged in zero-distance from the sample surface. The newly developed PCB-H-coils prevail the widespread wire-wound coils in indisputable ways, mainly due to their incompressibility. We present here a coil design that fulfils the corresponding demands in acceptable ways. This is attained by an incompressible H-coil of minimum thickness, extreme coil area and elastic on-press to the sample surface.
We present a first-principles study of Cr-capped Co5/Pd(111) ultrathin films in the near-compensated regime where Co stacking, structural relaxation, and Cr–Co exchange coupling compete on comparable energy scales and jointly determine the magnetic anisotropy. Three representative Co growth stackings are considered: pure fcc, hcp-rich, and mixed fcc+hcp. The hcp-rich stacking is found to be the lowest-energy structure and the most favorable configuration for approaching perpendicular magnetic anisotropy (PMA). We find that the magnetocrystalline anisotropy (MCA) of the non-relaxed films remains in-plane, whereas structural relaxation strongly suppresses the in-plane anisotropy and brings the system close to the spin-reorientation transition (SRT). Within PBE and the magnetic-force-theorem approach, antiferromagnetic Cr–Co coupling shifts the anisotropy toward the perpendicular side and yields the largest positive MCA for the relaxed hcp-rich film. The inclusion of moderate on-site Coulomb corrections through the Ueff parameter further shows that the hcp-rich AF configuration remains the structural ground state, but reveals that the Cr–Co exchange-energy scale and the precise balance between magnetocrystalline and shape anisotropy are Ueff-dependent. Within PBE, inclusion of the dipolar shape-anisotropy term keeps all configurations effectively in-plane. More generally, the results identify Cr/Co5/Pd(111) as a correlation-sensitive system close to a spin-reorientation boundary, rather than as either a universally in-plane film or a robust PMA state. Layer- and k-resolved analyses reveal that this near compensation originates from a competition between buried and middle Co layers, which favor in-plane magnetization, and the upper Pd region, the top Co layer beneath Cr, and the Cr cap, which favor the out-of-plane direction. In reciprocal space, the MCA is governed by the cancellation of positive and negative finite-k hot spots rather than by states near the Γ-point. Fully self-consistent spin–orbit calculations show that the relaxed AF configurations are especially fragile: the frozen-potential approximation overestimates the tendency toward perpendicularity and can even reverse the direction of the easy-axis. Our results identify the relaxed hcp-rich AF Cr/Co5/Pd(111) film as the closest precursor to robust PMA and establish stacking, relaxation, and interfacial exchange as key control parameters for tuning Cr/Co/Pd-based films across the SRT.
Unsupervised image clustering remains challenging because visually similar categories often lack explicit semantic boundaries, while purely visual representations are easily affected by appearance variations, background noise, and ambiguous local structures. Although recent image–text clustering methods introduce external textual semantics through vision–language models and lexical knowledge bases, their contrastive optimization still largely depends on static visual views and manually designed augmentations, which may provide insufficient or noisy positive samples. To address these limitations, we propose Diffusion-Augmented Adaptive Image–Text Contrastive Learning (DAITC), a unified unsupervised multimodal clustering framework that integrates adaptive text counterpart construction with diffusion-based semantic positive generation. Specifically, CLIP is first used to extract image and text embeddings, and a compact semantic vocabulary is constructed from WordNet according to image-level semantic centers. An adaptive temperature mechanism is then introduced to generate image-specific text counterparts by dynamically weighting candidate noun embeddings according to their similarity distributions. Beyond conventional visual augmentation, we further design a text-guided conditional diffusion module that generates semantically consistent positive views conditioned on both visual embeddings and adaptive text counterparts. These diffusion-augmented samples are incorporated into a reliability-aware soft contrastive objective, together with direct image–text alignment and neighborhood-level cross-modal consistency constraints. Finally, cluster balance and confidence regularization are employed to obtain stable and discriminative cluster assignments. Experiments are conducted on STL-10, CIFAR-10, CIFAR-20, DTD, and UCF-101 to evaluate clustering accuracy, semantic alignment, robustness, and generalization ability. The proposed framework provides a reliable way to combine external textual semantics and generative positive augmentation for unsupervised multimodal image clustering.
Magnetic levitation provides a contactless suspension mechanism for precision measurement and laboratory testing systems where mechanical friction and support interference must be minimised. However, the disturbance response of compact multi-coil maglev platforms remains insufficiently understood. This study experimentally investigates a four-coil electromagnetic levitation platform, focusing on multi-axis suspension behaviour and magnetic force redistribution under external perturbations. Controlled pitch/ roll and yaw disturbances were applied while coil current responses, platform displacement and steady vibration amplitude were measured. The disturbance moment was increased to 0.0172 N·m. Under pitch/roll disturbance, the measured air gap on the disturbed side increased from 24.8 to 26.9 mm, whereas the yaw disturbance produced a smaller air gap change from 24.8 to 25.6 mm. Pitch/ roll disturbances generated greater coil current redistribution and vibration response because they directly modified the vertical suspension force balance, while yaw disturbances were affected mainly by weaker lateral magnetic coupling. The influence of platform mass and nominal air gap on suspension sensitivity is further investigated, demonstrating that reduced air gaps and increased the total weight enhance disturbance sensitivity. These results provide quantitative guidance for the design and performance assessment of magnetic levitation platforms.
