LiDAR-Guided Illumination-Aware 3D Gaussian Splatting for Cultural Heritage

UH-GS: Uncertainty-aware hierarchical Gaussian splatting for outdoor scene reconstruction

Abstract. In outdoor scene reconstruction, dynamic occlusions and multiscale structures often undermine multiview consistency and hinder effective gradient accumulation of high-frequency Gaussian primitives, leading to artifacts and the loss of fine details in Gaussian splatting–based radiance field methods. To address these challenges, we propose an uncertainty-aware hierarchical Gaussian splatting framework for outdoor 3D reconstruction. Specifically, our method constructs a hierarchical octree-based spatial representation from the results of aerial triangulation. It introduces level of detail constraints to enable structured management and progressive optimization of Gaussian primitives across different scales. This design effectively alleviates the imbalance in training and the redundant growth of Gaussian primitives commonly observed in multiscale outdoor scenes. In addition, we incorporate an uncertainty prediction mechanism that evaluates the consistency between rendered results and ground-truth images in the feature space, allowing the model to automatically identify dynamically occluded regions and suppress their gradient contributions during optimization. As a result, the adverse impact of dynamic artifacts on static scene modeling is substantially reduced. Experimental results demonstrate that, without incurring significant additional training overhead, our method consistently improves structural consistency and fine-detail reconstruction quality in outdoor scenes, while simultaneously reducing model complexity and maintaining real-time rendering performance. Furthermore, the proposed approach can be seamlessly integrated into multiple mainstream Gaussian splatting frameworks, exhibiting strong robustness and promising potential for practical deployment.

3D Reconstruction Technology of Virtual Art Museum Integrating Neural Radiation Field

High-fidelity three-dimensional reconstruction of virtual art museums is essential for digital cultural heritage preservation and immersive visual communication, while also providing valuable references for computational imaging and electromagnetic-based scene perception under complex radiance propagation conditions. However, conventional Neural Radiance Field (NeRF) methods are limited by unstable training and texture distortion caused by specular reflections, translucent materials, and non-uniform illumination, making it difficult to preserve fine artistic details. To address these challenges, this study proposes a Distortion-Aware Ray Sampling Neural Radiance Field (DARS-NeRF) framework that integrates distortion-aware ray sampling with structure–material–lighting decoupled modeling. The proposed method enhances sampling density in visually distorted regions while combining Mask2Former semantic segmentation and a Light Estimation Network to provide semantic priors and illumination constraints for physically consistent reconstruction. Experiments conducted on 14,700 real-world art gallery images demonstrate that DARS-NeRF achieves PSNR, SSIM, and LPIPS values of 28.9 dB, 0.912, and 0.142, respectively, outperforming the strongest baseline by 2.2 dB, 3.2%, and 16.0%. In addition, Albedo MAE and Roughness RMSE are reduced by 18.3% and 20.4%, while View Consistency reaches 6.42 lux, indicating superior robustness to illumination variations. The proposed framework enables collaborative reconstruction of geometry, materials, and lighting with enhanced physical consistency, providing an effective paradigm for virtual museums, digital preservation, remote education, and high-fidelity intelligent visual sensing.

Evaluating 3D Scene Representations for Aerial Photogrammetry across Diverse Cityscapes

Abstract. The proliferation of continuous Neural Radiance Field (NeRF) and 3D Gaussian Splatting (3DGS) has shifted the paradigm of 3D aerial reconstruction from relying solely on geometric stereo matching to inverse rendering optimization. However, while these emerging rendering-based frameworks excel in synthesizing photo-realistic novel views, their capability to extract accurate surfaces in complex aerial scenarios remains ambiguous compared to traditional methods. To establish a clearer understanding, this study presents a comprehensive evaluation of five representative frameworks spanning traditional Structure from Motion (SfM), purely Signed Distance Field (SDF) representations, unstructured 3D Gaussians, hybrid voxel-Gaussians, and strictly explicit sparse voxels. By systematically standardizing identical computational environments, inputs, and unified mesh-extraction pipelines on both real-world airborne LiDAR datasets and synthetic cityscapes, we assess their performance regarding visual fidelity, geometric accuracy, and resource efficiency. The experimental results reveal that while traditional MVS produces the highest overall geometric precision by strictly enforcing multi-view rigid geometry, it is prone to failures in texture-less regions. Among rendering-based representations, a fundamental trade-off exists: highly flexible, unstructured 3DGS achieve highest visual scores but degrade the underlying geometric surfaces; conversely, explicitly structured techniques demonstrate distinct superiority in regularizing topological coherence and floating artifact suppression. Furthermore, we observe that integrating structured voxels avoids the severe memory bottlenecks associated with extracting geometries from chaotic unorganized primitives. These empirical findings emphasize that for large-scale aerial photogrammetry, integrating explicit spatial structuralization into differentiable rendering pipelines is imperative for achieving scalable operations and bridging the geometric accuracy gap with traditional methods.

High-Fidelity Gaussian Splatting from MVS Clouds: An Iterative Spatial Decomposition framework

This work proposes an Iterative Spatial Decomposition framework that bridges dense geometric priors from Multi-View Stereo (MVS) with Gaussian Splatting and introduces Hierarchical Geometric Prior Sampling (HGPS), which substantially reduce redundancy in MVS point clouds while preserving critical details, thereby providing a more robust geometric foundation for reconstruction.

Edge-aware gaussian splatting: enhancing boundary fidelity in 3D scene rendering

Geometry-Semantics Co-Regularization for Gaussian Splatting in Indoor Reconstruction.

A geometry-semantics co-regularization framework that jointly optimizes geometry and semantics within 3DGS and develops a multi-view semantic consistency supervision to regularize the semantic distributions of Gaussian primitives, ensuring cross-view consistency for Gaussians corresponding to the same semantic category or instance.