ParticleGS is a visualization-aware framework based on 3D Gaussian Splatting (3DGS) that learns a compact representation directly optimized for rendered image quality, combining a multi-stage, multi-orbit training pipeline and a lightweight network that adapts a single trained model to user-specified visualization parameters at inference time.
Abstract
Large-scale particle simulations produce hundreds of millions of particles, straining storage, transfer, and interactive visualization. Existing lossy compressors such as SZ3 operate in data space and provide no guarantees on downstream visualization fidelity. We propose ParticleGS, a visualization-aware framework based on 3D Gaussian Splatting (3DGS) that learns a compact representation directly optimized for rendered image quality, combining (1) a multi-stage, multi-orbit training pipeline, (2) VizMapper, a lightweight network that adapts a single trained model to user-specified visualization parameters at inference time, and (3) spatial block training with KD-tree decomposition and global fine-tuning. On a 281-million-particle HACC cosmological simulation, our 8-block model reaches 30.03 dB PSNR at 65x compression, outperforming SZ3 by 5-8 dB at comparable ratios, and generalizes without tuning to additional HACC regions and a dark-matter-only FIRE-2 simulation. It renders at 662 FPS on a single GPU, over 2,300x faster than ParaView on the full particle data.
The proposed framework provides a simple and effective path toward fast and inherently compact 3DGS training, and leverages a synergistic design: an L2 reconstruction loss to provide error-proportional gradients that stabilize optimization, and a novel Polarized Opacity Prior to actively manage the Gaussian population.
Zi-Ming Wang, Kai-Wen Duan, Ko-Wei Huang et al.· 0 citations
Point cloud is a crucial 3D representation that plays significant role in fields such as VR/AR and digital museums. With the increasing amount of point cloud data, many compression and rendering methods are proposed, which often focus on either 3D or 2D visual quality. However, with the rapid advancements in display technology, there is a growing expectation for improved 2D and 3D visual quality while using lower bandwidth. To address these challenges, we present PCCRender, a novel end-to-end framework that achieves high-quality attribute preservation and rendering capabilities while maintaining low bandwidth requirements. To achieve efficient variable rate compression, we introduce Geometry-Invariant Rate Adjustment (GIRA) module that mitigates the influence of point cloud density during rate adjustment. Additionally, to improve decoding speed, we develop Uneven Four-Group Context Model (UFCM), achieving a trade-off between the accuracy and complexity of entropy parameter prediction. Moreover, we design Voxel to Gaussian Primitive Converter (V2G-Converter) which generates Gaussian primitives from decoded point clouds, enabling differentiable rendering. Our unified optimization framework jointly minimizes bit rate while maximizing both 2D rendered image quality and point cloud attribute fidelity. Experimental results demonstrate that PCCRender achieves state-of-the-art performance. Compared to the GPCC v23 (GS Render) method, our framework achieves 10.58% BD-Rate gain in attribute compression and 1.5 dB BD-PSNR increase in 2D rendered visual quality. These compelling results, combined with support for variable rate and free viewpoint rendering, establish PCCRender as a practical solution for the next generation point cloud applications.
Kangli Wang, Wei Cheng, Ronggang Wang et al.· IEEE Transactions on Image P...· 0 citations
InfoLoD introduces a Fisher-guided self-distillation scheme that uses the Fisher Information Matrix to select geometrically valid, information-rich pseudo viewpoints, enabling LoD training directly from a pre-trained 3DGS model without any original images.
Zhenyu Xia, Pengcheng Han, Lin Chen et al.· IEEE Transactions on Visuali...· 0 citations
In scientific simulation, regular grids, unstructured meshes, and particle-based formats are chosen to represent field data for computational efficiency, geometry/adaptive flexibility, and following motion/deformation, respectively. Each of these field data formats is often handled through separate data-specific processing pipelines. We present a unified sample-based Gaussian encoding method that represents these data forms under a single fixed-budget formulation. The method initializes and refines Gaussian primitives directly from the input samples while preserving a prescribed primitive count and encoded size to achieve a desired level of data reduction. Across structured, unstructured, and particle data, the sample-based formulation improves reconstruction accuracy with measurably fewer primitives in comparison to prior formulations, achieving up to 4.8 dB higher PSNR with an approximate 44x reduction in primitive count. For time-varying data, warm-starting from the previous timestep reduces the optimization required to reach independently trained reconstruction quality. Together, these results demonstrate a unified fixed-budget Gaussian encoding framework for structured, particle, unstructured, and time-varying scientific data with predictable storage, higher reconstruction accuracy, and improved temporal encoding efficiency.
Michael R. Martin, J. Insley, Victor A. Mateevitsi et al.· 0 citations
Recent advances in 3D content generation have demonstrated the effectiveness of optimization-based frameworks such as DreamGaussian, which combine 3D Gaussian Splatting (3DGS) with diffusion-guided Score Distillation Sampling (SDS) to efficiently synthesize 3D assets from a single image. Compared with earlier NeRF-based approaches, these methods provide significantly faster training and rendering while maintaining competitive visual quality. However, existing approaches still suffer from several limitations, including floating artifacts, unstable geometry reconstruction, noisy or over-smoothed textures, and inconsistent appearance across viewpoints. These issues mainly arise from the absence of explicit geometric and appearance constraints during optimization, resulting in reduced visual fidelity and multi-view consistency. To address these limitations, we propose two lightweight yet effective regularization modules for enhancing single-image 3DGS-based image-to-3D generation. We introduce an entropy-based alpha regularization that minimizes the binary entropy of rendered opacity maps, encouraging more confident occupancy predictions and suppressing ambiguous semi-transparent regions. This regularization effectively reduces floating artifacts and improves CLIP-Similarity by 1.81% compared with DreamGaussian. In addition, the proposed normal-guided texture smoothness loss enhances texture coherence, yielding an additional 0.41% improvement. By combining both regularization modules, our method achieves an overall performance gain of 2.26% over the baseline. These findings suggest that incorporating simple geometric and appearance priors can substantially enhance the robustness and fidelity of diffusion-guided 3D Gaussian Splatting for high-quality image-to-3D generation. Code is available at https://github.com/NamQuanProject/ENEA-GS
3D Gaussian Splatting (3DGS) enables real-time rendering of photorealistic scene representations from multiview images and has potential as a visualization layer for digital twins. However, photorealistic appearance and image-level metrics alone do not establish practical suitability. We evaluate standard 3DGS on six real-world datasets: wall, forest, piled-pier underside, steel-girder bridge, asphalt pavement, and outdoor sculpture. The framework combines held-out test-view metrics (PSNR, SSIM, and LPIPS), spatial error maps, region-of-interest (ROI) inspection, computational records, scale–opacity grouping, and Gaussian-to-MVS distance analysis. In the wall case study, extending adaptive density control from 7000 to 30,000 iterations nearly doubled Gaussian count and model size and increased training time by about 50%, with only modest test-view improvements. Scale–opacity classes did not reliably distinguish high- from low-error regions. Gaussian-to-MVS distance was more strongly associated with scale than learned opacity, and median local errors generally increased with distance, although distributions overlapped substantially. Across datasets, major structures were generally reproduced well, whereas sparse foliage, low-contrast repetitive textures, fine surface details, and distant background objects required ROI-level verification. The best test-view performance reached 37.11 dB PSNR, 0.952 SSIM, and 0.193 LPIPS. Practical evaluation should consider global quality together with local, computational, and geometric characteristics.
Tomohiro Mizoguchi· Italian National Conference...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
