Abstract. This study investigates the use of synthetic images generated within Blender for reconstruction via 3D Gaussian Splatting (3DGS). These synthetic images are derived from a 4D parametric model of a Rhenish castle, incorporating its surroundings and distant environment. While such parametric models offer high-fidelity data, they are computationally intensive for real-time applications. 3DGS is therefore employed to produce high-quality visualisations from images with known spatial orientations. Two reconstruction methods are compared in this study: the open-source native code and the commercial Postshot solution with its Splat3 model. The primary objective is to demonstrate the applicability of this method using synthetic imagery to create lightweight visualisations of digital twins of theoretical 4D states. The underlying parametric model, comprising numerous distinct objects and procedural textures, achieves high photorealism at the expense of substantial computational resources. Consequently, the reconstruction of this dataset via 3DGS facilitates the export and online dissemination of the complex model, decoupling visualisation quality from geometric complexity. The approach is quantitatively validated by comparing the 3DGS output against the original ground truth. Results demonstrate that the Splat3 model outperforms the native open-source approach in visual fidelity, processing speed, and geometric accuracy when handling high-resolution datasets. Both reconstruction methods achieve rendering performances well above 100 frames per second. This confirms that 3DGS can successfully be used with synthetic images to transform computationally heavy parametric models into highly optimised digital representations, ensuring near real-time visualisation suitable for immersive virtual reality and public dissemination.
E. Sommer, Arnadi Murtiyoso, M. Koehl et al.· The International Archives o...· 0 citations
Abstract. Accurate 3D reconstruction of Cultural Heritage (CH) assets remains a challenging task when scenes include complex geometries and non-Lambertian surfaces, such as dense vegetation, reflective ceramics, or polished materials, which often degrade the performance of traditional multi-view stereo (MVS) pipelines. This work investigates the potential of Mesh-In-the-Loop Gaussian Splatting (MILo), a recent extension of 3D Gaussian Splatting (3DGS) that integrates differentiable mesh extraction directly within the optimization process, enabling bidirectional consistency between volumetric and surface representations. The method is evaluated on three challenging CH datasets: a monumental Tilia tomentosa tree located in a UNESCO-listed urban garden, a reflective ceramic object from the Sarreguemines Earthenware Museum and the South Portal façade of the Notre-Dame Cathedral of Strasbourg. MILo-based reconstructions are compared against standard photogrammetric MVS results generated with Agisoft Metashape, using terrestrial laser scanner (TLS) point clouds as geometric reference. Quantitative accuracy assessment is performed through Multiscale Model-to-Model Cloud Comparison (M3C2), focusing on error distribution, standard deviation, outlier percentage, and preservation of fine-scale structures. Results indicate that while conventional MVS performs slightly better on stable architectural surfaces, MILo significantly improves reconstruction consistency for complex organic geometries, substantially reducing outliers and better preserving thin structures. These findings highlight the suitability of MILo for CH documentation scenarios characterized by challenging surface properties and intricate natural forms.
D. Billi, Chaimaa Delasse, Arnadi Murtiyoso et al.· The International Archives o...· 0 citations
