publications

2026

ISPRS Archives
Automatic Segmentation of 3D Gaussian Splatting in Cultural Heritage Complex

Widiatmoko Azis Fadilah, Virgile Gauthier, Arnadi Murtiyoso, and 2 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

3D Gaussian Splatting (3DGS) has emerged as a promising method for photorealistic scene reconstructions, yet its application to semantic segmentation in real-world heritage documentation remains underexplored. This study proposes and evaluates an automated semantic 3DGS segmentation pipeline integrating the Segment Anything Model 3 (SAM 3) with per-class prompting for Gaussian reconstruction, applied to a nadiral UAV dataset of the Siti Inggil heritage complex in Cirebon, Indonesia. Segmentation performance of four semantic classes (ground, roofs, vegetations, and water bodies) were assessed against manually segmented 2D and 3D reference data, supplemented by geometric accuracy assessment via the M3C2 analysis. Results reveal both the promise and the inherent challenges of applying 3DGS segmentation to complex real-world heritage scenes, where acquisition geometry, surface characteristics, and foundational model limitations can be observed.
@article{fadilah2026automatic, title = {Automatic Segmentation of 3D Gaussian Splatting in Cultural Heritage Complex}, author = {Fadilah, Widiatmoko Azis and Gauthier, Virgile and Murtiyoso, Arnadi and Landes, Tania and Grussenmeyer, Pierre}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {49}, pages = {825--832}, year = {2026}, doi = {10.5194/isprs-archives-XLIX-B2-2026-825-2026}, dimensions = {true}, project = {semG}, publisher = {Copernicus Publications G{\"o}ttingen, Germany} }
ISPRS Archives
3D Meshing of Challenging Surfaces using Gaussian Splatting

Dario Billi, Chaimaa Delasse, Arnadi Murtiyoso, and 4 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

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.
@article{billi20263d, title = {3D Meshing of Challenging Surfaces using Gaussian Splatting}, author = {Billi, Dario and Delasse, Chaimaa and Murtiyoso, Arnadi and Macher, H{\'e}l{\`e}ne and Grussenmeyer, Pierre and Caroti, Gabriella and Piemonte, Andrea}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {49}, pages = {801--808}, year = {2026}, doi = {10.5194/isprs-archives-XLIX-B2-2026-801-2026}, dimensions = {true}, project = {mesh}, publisher = {Copernicus GmbH} }
ISPRS Archives
In-Situ Gaussian Splatting-generated 3D Thermal Mesh Visualization for Urban Trees in Augmented Reality

Chaimaa Delasse, Dario Billi, I Gede Mahendra Darmawiguna, and 6 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

Urban trees provide critical ecosystem services in dense city environments, yet current workflows for monitoring their thermal behaviour remain confined to 2D desktop-based analysis with no three-dimensional spatial context or field-deployable visualization capability. This paper presents a complete pipeline for in-situ 3D thermal mesh visualization of urban trees in Augmented Reality (AR), combining Thermal InfraRed (TIR) image acquisition, Gaussian Splatting-based mesh reconstruction, quantitative validation, and mobile AR deployment. TIR images of a Tilia tomentosa acquired with a FLIR T560 camera are preprocessed with a standardized false-colour palette and fed into the MILo (Mesh-In-the-Loop Gaussian Splatting) framework to reconstruct a thermally attributed 3D mesh. Geometric evaluation against a Z+F IMAGER 5016 TLS reference using the M3C2 algorithm demonstrates that MILo recovers 13.5 times more canopy geometry than traditional multi-view stereo under thermal imagery, with a standard deviation of 4.0 cm. A colourmap inversion procedure recovers per-vertex temperature estimates from the GS-derived mesh colours, yielding a mean absolute difference of 0.7°C against direct T-Cam measurements (thermal camera mounted on the laser scanner), within the combined instrument accuracy of both sensors. The resulting thermal Gaussian Splat was deployed in a custom Android AR application supporting hybrid marker-based and GPS-based spatial anchoring for in-situ visualization. These results demonstrate the technical feasibility of GS-based thermal reconstruction and mobile AR as a medium for communicating three-dimensional canopy thermal information to educators and urban forestry practitioners.
@article{delasse2026situ, title = {In-Situ Gaussian Splatting-generated 3D Thermal Mesh Visualization for Urban Trees in Augmented Reality}, author = {Delasse, Chaimaa and Billi, Dario and Darmawiguna, I Gede Mahendra and Satriadi, Kadek Ananta and Murtiyoso, Arnadi and Macher, H{\'e}l{\`e}ne and Lecomte, Vincent and Hajji, Rafika and Landes, Tania}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {49}, pages = {1021--1028}, year = {2026}, doi = {10.5194/isprs-archives-XLIX-B2-2026-1021-2026}, dimensions = {true}, project = {mesh}, publisher = {Copernicus Publications G{\"o}ttingen, Germany} }
ISPRS Archives
Empirical Assessment of Geometric Accuracy of Underwater Lidar in Tropical Shallow Waters

Mentari K Azzahra, Fickrie Muhammad, Arnadi Murtiyoso, and 3 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

This study empirically evaluates the geometric accuracy of point cloud data acquired using an underwater lidar (ULi) system in a tropical shallow water environment. The field test was conducted in the tropical waters of the Seribu Islands, Indonesia, characterised by relatively low turbidity. Terrestrial laser scanning (TLS) and close-range photogrammetry were employed as independent reference datasets. Geometric discrepancies between datasets were quantified using the multiscale model-to-model cloud comparison (M3C2) algorithm, and errors were statistically characterised using the median and median absolute deviation (MAD) to ensure robustness under non-normal distributions. The results indicate that the error of ULi relative to TLS is 0.008 ± 0.012 m, while the error relative to photogrammetry is 0.006 ± 0.013 m. In comparison, the discrepancy between photogrammetry and TLS is smaller, at 0.002 ± 0.004 m. Dimensional analysis of an acoustic Doppler current profiler (ADCP) frame further shows that ULi agrees with TLS and photogrammetry within the millimetre to centimetre range (0.000–0.015 m). Larger deviations in specific segments are attributed to local effects, including edge-related artefacts. Overall, the results demonstrate that ULi provides reliable geometric measurements in shallow-water conditions with low turbidity. Despite slightly lower accuracy compared to terrestrial methods, the system shows potential for underwater mapping applications, particularly in shallow water environments.
@article{azzahra2026empirical, title = {Empirical Assessment of Geometric Accuracy of Underwater Lidar in Tropical Shallow Waters}, author = {Azzahra, Mentari K and Muhammad, Fickrie and Murtiyoso, Arnadi and Scheider, Annette and Sternberg, Harald and Alodia, Gabriella}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {49}, pages = {923--930}, year = {2026}, doi = {10.5194/isprs-archives-XLIX-B2-2026-923-2026}, dimensions = {true}, publisher = {Copernicus Publications G{\"o}ttingen, Germany} }
ISPRS Archives
Gaussian splatting for the reconstruction of complex and highly detailed object

Sara Gonizzi Barsanti, Dario Billi, Etienne Sommer, and 2 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

In recent years, Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have emerged as advanced methods for photogrammetry-based 3D reconstruction. Since its introduction in 2020, NeRF has gained significant attention due to its capability to generate high-fidelity reconstructions from multi-view imagery. More recently, 3D Gaussian Splatting (3DGS), introduced in 2023, has proposed an alternative explicit scene representation based on a collection of anisotropic Gaussian primitives optimized directly in 3D space. This representation allows efficient rendering and scalable modelling of complex scenes while maintaining high visual quality. This paper analyses the performance of different 3DGS methods when dealing with complex geometry and less-cooperative surfaces compared to standard SfM IM procedures. Included in the comparison is also the Mesh-In-the-Loop Gaussian Splatting for Detailed and Efficient Surface Reconstruction (MILo), a novel meshing method using Gaussian splats. Three Gaussian splatting methods as implemented in the Postshot commercial software were also tested. Our experiments show that MILo shows very promising results in terms of detail reconstruction, while standard Gaussian splatting excels in visualisation but is still plagued by a high rate of noise especially when converted into a geometric point cloud form.
@article{gonizzi2026gaussian, title = {Gaussian splatting for the reconstruction of complex and highly detailed object}, author = {Gonizzi Barsanti, Sara and Billi, Dario and Sommer, Etienne and Elefante, Erika and Murtiyoso, Arnadi}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {49}, pages = {1355--1361}, year = {2026}, doi = {10.5194/isprs-archives-XLIX-B2-2026-1355-2026}, dimensions = {true}, project = {mesh}, publisher = {Copernicus GmbH} }
ISPRS Archives
A Hybrid Approach using Gaussian Splatting and Parametric Models based on 3D Renders for Real-Time Visualisation

Etienne Sommer, Arnadi Murtiyoso, Mathieu Koehl, and 1 more author

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

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.
@article{sommer2026hybrid, title = {A Hybrid Approach using Gaussian Splatting and Parametric Models based on 3D Renders for Real-Time Visualisation}, author = {Sommer, Etienne and Murtiyoso, Arnadi and Koehl, Mathieu and Grussenmeyer, Pierre}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {49}, pages = {551--558}, year = {2026}, doi = {10.5194/isprs-archives-XLIX-B2-2026-551-2026}, dimensions = {true}, project = {GS3D}, publisher = {Copernicus Publications G{\"o}ttingen, Germany} }
ISPRS Archives
Augmenting City Models with Handheld LiDAR and 3D Gaussian Splatting for Inclusive Pedestrian Infrastructure Assessment

Deni Suwardhi, Wahyunan Andika, Ratri Widyastuti, and 5 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

Urban digital twins increasingly require pedestrian-scale three-dimensional (3D) representations to support accessibility and inclusiveness assessment. However, existing approaches typically emphasize either geometric accuracy or visual realism, while lacking an integrated framework for analysing pedestrian-level conditions. This study proposes a hybrid workflow integrating handheld LiDAR and 3D Gaussian Splatting (3DGS) within a CityGML-based semantic framework for accessibility assessment. Handheld LiDAR provides centimetre-level geometric measurements, enabling the extraction of key indicators such as slope, surface roughness, and obstacle presence. In parallel, 3DGS reconstruction from 360° video imagery enhances visual realism and perceptual understanding. Both datasets are co-registered and structured within the CityGML 3.0 Transportation model to represent pedestrian environments in a unified spatial and semantic framework. Accessibility assessment was conducted using three approaches: LiDAR-based analysis, field survey observations, and immersive evaluation in a Virtual Reality (VR) environment. The LiDAR-based results were used as reference. Comparative analysis shows that field survey assessment achieves an agreement of approximately 85.7%, while VR-based assessment reaches approximately 75.4%. The results indicate that while VR does not replace metric-based analysis, it enables perception-driven and participatory evaluation. In particular, VR-based assessment shows potential to involve users, including people with disabilities, in accessibility evaluation through immersive and remote interaction. The proposed approach contributes to the development of human-scale urban digital twins by integrating metric accuracy, semantic structure, and participatory evaluation for more inclusive accessibility analysis.
@article{suwardhi2026augmenting, title = {Augmenting City Models with Handheld LiDAR and 3D Gaussian Splatting for Inclusive Pedestrian Infrastructure Assessment}, author = {Suwardhi, Deni and Andika, Wahyunan and Widyastuti, Ratri and Fadilah, Widiaomoko Azis and Murtiyoso, Arnadi and Grusennmeyer, Pierre and Remondino, Fabio and Helmy, Farhan}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {49}, pages = {1117--1124}, year = {2026}, doi = {10.5194/isprs-archives-XLIX-B2-2026-1117-2026}, dimensions = {true}, publisher = {Copernicus Publications G{\"o}ttingen, Germany} }
ISPRS Archives
Bridging Semantic Mesh, CityGML, and Gaussian Splatting for Urban Modelling and Visualization

Deni Suwardhi, Muhammad Arif Sudibyo, Agus Ambarwari, and 7 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

Urban digital twin systems require 3D city representations that reconcile semantic structure, geometric reliability, simulation capability, and photorealistic real-time rendering. Existing approaches usually prioritize a single modelling paradigm, limiting their ability to support both analytical and visualization needs. CityGML provides standardized semantics and topology but often lacks surface realism. Semantic mesh models preserve geometric detail suitable for environmental simulations but provide limited hierarchical semantics. In contrast, neural radiance-field approaches such as 3D Gaussian Splatting (3DGS) enable photorealistic rendering at interactive frame rates but do not explicitly encode topology or structured semantics. This study establishes a comparative framework linking LiDAR-derived CityGML, semantic mesh, 3D Gaussian Splatting, and Triangle Splatting within a unified urban modelling workflow. UAV data acquired using a DJI ZENMUSE L2 sensor serve as the geometric backbone for reconstructing CityGML LoD1–LoD2 models. The semantic model is transformed into a textured triangular mesh, while radiance-based models are generated from the same imagery using multiple 3DGS implementations and a triangle splatting framework. Comparative evaluation investigates geometric coherence, semantic preservation, and radiance consistency to identify structural correspondences across the representations. The results reveal complementary modelling layers that can be systematically mapped rather than treated as competing alternatives. Based on these findings, the paper proposes a conceptual foundation for a unified 3D urban model capable of transforming consistently into semantic, surface-based, and radiance-based representations for adaptive urban digital twin systems. Data are freely accessible for research purposes at https://github.com/3DOM-FBK/urban-representation-fusion/.
@article{suwardhi2026bridging, title = {Bridging Semantic Mesh, CityGML, and Gaussian Splatting for Urban Modelling and Visualization}, author = {Suwardhi, Deni and Sudibyo, Muhammad Arif and Ambarwari, Agus and Andika, Wahyunan and Ihsan, Muhammad and Widyastuti, Ratri and Fadilah, Widiatmoko Azis and Murtiyoso, Arnadi and Grussenmeyer, Pierre and Remondino, Fabio}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {49}, pages = {81--88}, year = {2026}, doi = {10.5194/isprs-archives-XLIX-B4-2026-81-2026}, dimensions = {true}, publisher = {Copernicus GmbH} }
ISPRS Archives
Exploring Point Transformers on 3D Semantic Segmentation of Javanese Architectures

T. Betsas, A. Murtiyoso, P. Grussenmeyer, and 1 more author

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

The complex geometry of Javanese architecture poses significant challenges for 3D semantic segmentation in cultural heritage documentation. This study evaluates state-of-the-art Point Transformers, i.e., PTv1, PTv2, PTv3, and LitePT, on the Sewu temple dataset, focusing on robustness and efficiency. While PTv1 and PTv2 achieve the highest Intersection-over-Union (mIoU 0.71), they incur high computational costs. Conversely, LitePT provides an optimal balance, delivering competitive results (0.69 mIoU) while being drastically faster. Furthermore, experiments with limited data reveal the significant benefits of transfer learning from European heritage datasets. We conclude that efficient Point Transformer architectures are promising for the automated understanding of complex non-European monuments.
@article{betsas2026exploring, title = {Exploring Point Transformers on 3D Semantic Segmentation of Javanese Architectures}, author = {Betsas, T. and Murtiyoso, A. and Grussenmeyer, P. and Georgopoulos, A.}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {48}, pages = {57--64}, year = {2026}, doi = {10.5194/isprs-archives-XLVIII-2-W12-2026-57-2026}, dimensions = {true}, project = {semG}, publisher = {Copernicus Publications G{\"o}ttingen, Germany} }
ISPRS Archives
Plane-Intersection Reconstruction of Integrated Point Clouds toward Property Valuation-Oriented 3D Building Models

R. Widyastuti, D. Suwardhi, A. Hernandi, and 3 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

Three-dimensional (3D) building models with a high level of geometric and semantic detail are increasingly required to support fiscal cadastral applications, such as property valuation and taxation. In dense urban environments, however, the availability of complete and high-quality 3D data remains limited, particularly for exterior wall information. This research proposes a rule-based 3D building reconstruction workflow using integrated Unmanned Aerial Vehicle-Light-Detection and Ranging (UAV-LiDAR) and terrestrial handheld scanner point clouds to generate building models with hanging roofs and explicit building elements. The proposed method combines region growing segmentation for roof and exterior wall extraction, planar surface reconstruction through plane fitting, and geometric primitive intersection to construct roof meshes and building envelopes. Roof geometry is reconstructed from planar roof components, while exterior wall points are used to adjust the building envelope and support hanging roof modelling. The resulting 3D building model is represented as separated elements, including roof, exterior walls, and floor surfaces, enabling element-based area calculation and cost estimation for fiscal cadastral purposes. Quantitative evaluation using point cloud-to-model distance analysis shows that the reconstruction accuracy falls within the commonly reported range for Level of Detail 2 (LoD2) building models, despite limitations caused by segmentation precision and parameter sensitivity of rule-based algorithms. The results demonstrate that the proposed approach provides a feasible solution for 3D building modelling in environments with incomplete data, while supporting transparent and assessment-oriented fiscal cadastral workflows.
@article{widyastuti2026plane, title = {Plane-Intersection Reconstruction of Integrated Point Clouds toward Property Valuation-Oriented 3D Building Models}, author = {Widyastuti, R. and Suwardhi, D. and Hernandi, A. and Siagian, D.A.P. and Murtiyoso, A. and Remondino, F.}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {48}, pages = {503--509}, year = {2026}, doi = {10.5194/isprs-archives-XLVIII-2-W12-2026-503-2026}, dimensions = {true}, publisher = {Copernicus GmbH} }
ISPRS Archives
Metric Assessment of 3D Gaussian Splatting for UAV-Based Urban Heritage Reconstruction

W.A. Fadilah, A. Murtiyoso, T. Landes, and 1 more author

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

Urban heritage increasingly faces threats from natural disasters and anthropogenic activities, demanding preservation techniques that balance affordability, speed, and accuracy. Real-time rendering with 3D Gaussian Splatting (3DGS) offers immense potential for rapid heritage documentation; however optimal training configurations especially for aerial nadiral images remain limited. This study compares three 3DGS implementations: native 3DGS, Postshot Splat3, and MCMC variants against conventional multi-view stereo using an 80-meter altitude drone aerial imagery. Results reveal some interesting key findings where moderate downsampling of 1/8 resolution can substantially outperform full resolution training. Postshot Splat3 achieved optimal photometric quality while training 21x faster than native implementations. However, geometric analysis showed that rendering quality and geometric accuracy are not correlated. Furthermore, the result suggested that MCMC configurations failed to compete with native 3DGS and Postshot Splat3. These findings suggest that 3DGS-based heritage documentation requires a rather conservative resolution selection and optimized implementation for efficiency to allow cost effective preservation while maintaining dimensional fidelity.
@article{fadilah2026metric, title = {Metric Assessment of 3D Gaussian Splatting for UAV-Based Urban Heritage Reconstruction}, author = {Fadilah, W.A. and Murtiyoso, A. and Landes, T. and Grussenmeyer, P.}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {48}, pages = {143--150}, year = {2026}, doi = {10.5194/isprs-archives-XLVIII-2-W12-2026-143-2026}, dimensions = {true}, project = {semG}, publisher = {Copernicus Publications G{\"o}ttingen, Germany} }
ISPRS Archives
Integrating Terrestrial Laser Scanning and 3D Gaussian Splatting for Heritage Building Documentation and Monitoring

D. Suwardhi, L.K. Agustina, W. Andika, and 4 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2026

Abs DOI Bib

Accurate and reliable 3D documentation is important for the conservation and long-term monitoring of heritage buildings, where both metric precision and visual interpretability are required. This paper presents a co-registered hybrid 3D information model for heritage buildings, integrating Terrestrial Laser Scanning (TLS), Historic Building Information Modelling (HBIM), and 3D Gaussian Splatting (3DGS). The approach is demonstrated through a case study of Aula Timur ITB, a nationally protected heritage building in Indonesia. TLS data serve as the geometric backbone of the framework, providing survey-grade spatial reference. An HBIM model is generated from the TLS point clouds and structured using the Industry Foundation Classes (IFC) standard, ensuring that semantic building components are co-registered with the metric geometry. In parallel, a photorealistic 3DGS representation is generated from image-based reconstruction and rigidly aligned to the same TLS reference frame. Rather than producing a single fused geometry, the proposed integration results in a multi-layer representation in which metric, semantic, and radiance-field-based models coexist within a shared spatial reference. Quantitative alignment metrics, including transformation residuals, and cloud-to-cloud distance analysis, are used to assess geometric consistency between TLS and 3DGS. The integrated model is deployed in an immersive VR/MR environment to support intuitive navigation and visual inspection tasks. The results demonstrate that anchoring 3DGS to survey-grade geometry enables photorealistic real-time visualization while preserving spatial reliability, making the approach suitable for heritage documentation and monitoring scenarios. The proposed workflow highlights the complementary roles of metric surveying, information modelling, and emerging radiance-field representations in heritage-oriented 3D information systems.
@article{suwardhi2026integrating, title = {Integrating Terrestrial Laser Scanning and 3D Gaussian Splatting for Heritage Building Documentation and Monitoring}, author = {Suwardhi, D. and Agustina, L.K. and Andika, W. and Widyastuti, R. and Fadilah, W.A. and Murtiyoso, A. and Remondino, F.}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {48}, pages = {463--470}, year = {2026}, doi = {10.5194/isprs-archives-XLVIII-2-W12-2026-463-2026}, dimensions = {true}, publisher = {Copernicus Publications G{\"o}ttingen, Germany} }
CFS
Trends and Developments in the Realm of Virtual Forest Visualization

S.K.P. Kushwaha, H. Hristova, C. Fol, and 9 more authors

Current Forestry Reports, 2026

Abs DOI Bib

We examine key trends, developments, and challenges according to the global research in the use of 2D and 3D visualization for virtual forest experiences and present selected case studies. We describe the potential of virtual forest visualization for applications in forest management, conservation, education, and research. We also present an opportunity to utilize and develop further the scope of visualization using these latest technologies in forestry globally.
@article{kushwaha2026trends, title = {Trends and Developments in the Realm of Virtual Forest Visualization}, author = {Kushwaha, S.K.P. and Hristova, H. and Fol, C. and Hands, T. and Murtiyoso, A. and Coltekin, A. and Fedosov, A. and Holm, S. and Costa, M. and Bugmann, H. and Griess, V.C. and Schweier, J.}, journal = {Current Forestry Reports}, volume = {12}, number = {1}, pages = {3}, year = {2026}, doi = {10.1007/s40725-025-00265-5}, dimensions = {true}, publisher = {Springer International Publishing} }

2025

ISPRS Archives
Cross-domain urban and heritage point cloud benchmarks for semantic segmentation: insights from the 3M (Mapping Mappano Municipality) and ArCHX datasets

F. Matrone, A. Lingua, A. Murtiyoso, and 7 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2025

Abs DOI Bib

Semantic segmentation of 3D point clouds is a key enabler for intelligent urban analysis and cultural heritage documentation. However, current benchmarks remain uneven for different reasons: from the fragmented scenes of cityscape scenarios where a unique environment is not provided, to the difficulties in representing the great variety of architectural styles and lexicons in the cultural heritage (CH) domain. This study introduces two complementary datasets, the 3M (Mapping Mappano Municipality) dataset and the ArCHX (Architectural Cultural Heritage eXpanded) dataset, meant to enhance cross-domain learning between urban and architectural heritage environments. The 3M dataset, derived from UAV photogrammetry and GNSS-based surveys over an 11.9 km² area near Turin (Italy), provides a highly accurate, semantically labelled urban point cloud ( 900 M points) suitable for urban planning, automatic mapping, and infrastructure design. In parallel, the ArCHX dataset extends the existing ArCH benchmark with new scenes from Portuguese masonry gates and Indian Nagara-style temples, reaching over 400 million points. These additions diversify architectural typologies and improve the generalisation capacity of AI models applied to heritage semantic segmentation tasks. Methodological challenges in harmonising class taxonomies, managing heterogeneous materials, and aligning cultural architectural semantics across datasets are eventually discussed. Collectively, the 3M and ArCHX datasets establish a novel foundation for cross-domain benchmarking in 3D semantic segmentation, advancing data-driven research in digital twins, mapping, Historic Building Information Modelling (HBIM), and AI-assisted heritage documentation.
@article{matrone2025cross, title = {Cross-domain urban and heritage point cloud benchmarks for semantic segmentation: insights from the 3M (Mapping Mappano Municipality) and ArCHX datasets}, author = {Matrone, F. and Lingua, A. and Murtiyoso, A. and Sridhar, M. and Sanchez Aparicio, L.J. and Santamaria Maestro, R. and Sanz-Honrado, P. and Pande, H. and Seth Tiwari, P. and Agrawal, S.}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {XLVIII-1/W6-2025}, pages = {131--137}, year = {2025}, doi = {10.5194/isprs-archives-XLVIII-1-W6-2025-131-2025}, dimensions = {true}, publisher = {Copernicus}, }
IEEE Xplore
3D Gaussian Splatting for Archival of Balinese Temples from Community-Sourced Videos

A. Wilson, A. Murtiyoso, B. Jenny, and 5 more authors

In 2025 IEEE International Conference on Imaging Systems and Techniques (IST), 2025

Abs DOI Bib

We present a novel application of 3D Gaussian Splatting (3DGS) for the large-scale archival of thousands of temples in Bali. Our project, the Bali Digital Heritage Initiative, adopts a participatory approach in which local communities collect video footage of these vulnerable temples, which we transform into 3DGS scenes to serve as digital archives. Unlike existing approaches that require uploading curated sets of images, video contribution simplifies the process for the community. We expect this to encourage a high participation rate and increase the number of archived temples. However, these community-sourced videos are often of variable quality and usually aren’t ideal for 3D reconstruction, introducing technical challenges. This paper first presents several key challenges in the development of the 3DGS pipeline in this context, including managing video quality, combining multiple videos, ensuring reliable results from Structure from Motion, and maintaining the visual quality of 3DGS scenes. We then propose and test solutions to many of these challenges, addressing all parts of the reconstruction pipeline. We also introduce a postprocessing stage that cleans up the scene through a series of purpose-built 3DGS filters, which consider attributes of Gaussian primitives, including local density, geometry, and colors. Finally, we discuss future research directions to address unsolved problems and improve the performance of the pipeline.
@inproceedings{wilson20253d, title = {3D Gaussian Splatting for Archival of Balinese Temples from Community-Sourced Videos}, author = {Wilson, A. and Murtiyoso, A. and Jenny, B. and Darmawiguna, I.G.M. and Suputra, P.H. and Chandler, T. and Kesiman, M.W.A. and Satriadi, K.A.}, booktitle = {2025 IEEE International Conference on Imaging Systems and Techniques (IST)}, pages = {1--6}, year = {2025}, doi = {10.1109/IST66504.2025.11268436}, dimensions = {true}, publisher = {IEEE}, project = {GS3D} }
IEEE Xplore
Radiance Fields for Archaeological Documentation of Medieval Rhine Valley Castle Ruins

E. Sommer, A. Murtiyoso, M. Koehl, and 1 more author

In 2025 IEEE International Conference on Imaging Systems and Techniques (IST), 2025

Abs DOI Bib

This study investigates the use of Gaussian Splatting for 3D reconstruction of medieval castle ruins in the Rhine Valley, comparing its effectiveness with conventional photogrammetry. 3D Gaussian Splatting (3DGS), a recent real-time rendering technique, models scenes using point-based representations with anisotropic Gaussians, enabling high-fidelity reconstructions from sparse image inputs. Unlike traditional photogrammetry, which depends on dense point clouds and mesh generation, 3DGS offers smoother rendering, better handling of fine details, and improved performance in visually complex or degraded areas. 3DGS excels in producing visually compelling, immersive models with fewer artifacts in occluded or texture-deficient regions. In this paper, three analyses were performed by focusing on three individual aspects: processing time, geometric accuracy, and visual quality. Our study’s results highlight the advantages of 3DGS in this regard, which include quick data processing time and excellent visualisation capability specifically with finely detailed objects. However, this method still faces challenges when employed for a metric archiving of built heritage, where traditional photogrammetry still provides a higher quality metric result using the same input data. 3DGS suffers particularly from noisy data, especially when converted into point clouds. Nevertheless, this approach presents a promising tool for digital heritage preservation, enabling efficient and realistic visualisation of fragile historical sites.
@inproceedings{sommer2025radiance, title = {Radiance Fields for Archaeological Documentation of Medieval Rhine Valley Castle Ruins}, author = {Sommer, E. and Murtiyoso, A. and Koehl, M. and Grussenmeyer, P.}, booktitle = {2025 IEEE International Conference on Imaging Systems and Techniques (IST)}, pages = {1--6}, year = {2025}, doi = {10.1109/IST66504.2025.11268408}, dimensions = {true}, publisher = {IEEE}, project = {GS3D} }
ISPRS Archives
Integrative AI for the Understanding of Ancient Javanese Architectures

A. Murtiyoso, G. Mazzacca, F. Remondino, and 1 more author

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2025

Abs DOI Bib PDF

The use of digital techniques has seen an increasing amount of use in recent years for heritage documentation. The development of artificial intelligence (AI) also contributed to this rise, with many different applications to help facilitate the heritage recording process. A by-product of these developments is the increasing amount of available data, in tandem with the ever-increasing need for training data for AI purposes. This paper aims to re-use old datasets and repurpose them using modern methods. The objective is therefore to see if older datasets may be used to improve the quality of AI-based methods, while also investigating the use of new technologies such as Visual Language Models (VLM) to perform semantic queries and Gaussian splatting on these datasets. For this purpose, datasets from a previous documentation project involving Javanese “candi” architecture is used in this paper since this particular subject has not seen too many AI-based documentation research in the literature and is thus an interesting example to evaluate the generalisation of AI methods. Results show that old datasets can very well be used with modern techniques with promising results. In terms of semantic segmentation, machine learning yielded an overall accuracy of 0.89 while deep learning yielded 0.79. Several interesting inferences were also observed in the VLM query results, while Gaussian splatting showed very strong potential for visualisation-based applications to further enhance the reusability of these old datasets.
@article{isprs-archives-XLVIII-M-9-2025-1065-2025, author = {Murtiyoso, A. and Mazzacca, G. and Remondino, F. and Suwardhi, D.}, title = {Integrative AI for the Understanding of Ancient Javanese Architectures}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {XLVIII-M-9-2025}, year = {2025}, pages = {1065--1072}, doi = {10.5194/isprs-archives-XLVIII-M-9-2025-1065-2025}, dimensions = {true}, publisher = {Copernicus}, }
ISPRS Archives
Gaussian Splatting for Facade Orthophoto Generation – Comparison with MVS and TLS

A. Murtiyoso and H. Macher

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2025

Abs DOI Bib PDF

The notion of radiance fields is a very recent topic which has gained much traction in the past couple of years. Among some of the well known techniques, 3D Gaussian splatting (3DGS) seems to generate promising results in terms of 3D reconstruction. In the domain of heritage documentation, orthophotos are a standard product usually generated from photogrammetry by projecting image pixels into a 3D surface created from depth maps or point clouds. In this paper we investigate the possibility to substitute the conventional method for 3D surface generation, either terrestrial laser scanning (TLS) or multi-view stereo (MVS), with 3DGS. The paper will attempt to answer two questions: (1) whether this novel method is a viable solution for orthophoto generation within the heritage documentation context, and (2) at what point is 3DGS worth the time and effort to reach an acceptable orthophoto when compared to conventional methods. To do this, we perform multiple analysis on both the geometric quality of point clouds and the orthophotos themselves, using conventional TLS data as our reference. Our results indicate that 3DGS is not only promising, but indeed a viable alternative to conventional MVS with a potential gain of processing time between 12% up to 33% to reach a comparable quality. In the scenarios tested, both of these gains came with an average point cloud standard deviation of 2 cm.
@article{isprs-archives-XLVIII-M-9-2025-1059-2025, author = {Murtiyoso, A. and Macher, H.}, title = {Gaussian Splatting for Facade Orthophoto Generation – Comparison with MVS and TLS}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {XLVIII-M-9-2025}, year = {2025}, pages = {1059--1064}, doi = {10.5194/isprs-archives-XLVIII-M-9-2025-1059-2025}, dimensions = {true}, publisher = {Copernicus}, project = {GS3D}, }
ISPRS Archives
Initial Experiments on the Use of Radiance Fields for Underwater 3D Reconstruction

B. Tanduo, F. Matrone, and A. Murtiyoso

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2025

Abs DOI Bib PDF

Underwater photogrammetry presents unique challenges, including light attenuation, refraction, and turbidity, that affect the accuracy and quality of 3D reconstructions. This study investigates the performance of novel neural rendering techniques, Neural Radiance Fields (NeRF), SeaThru-NeRF, and 3D Gaussian Splatting (3DGS), in comparison to conventional Structure-from-Motion (SfM) workflows. Using a dataset acquired during the SIFET benchmark campaign on a submerged Roman archaeological site, we processed image data via Nerfacto, SeaThru, and Jawset Postshot (3DGS) and compared outputs against a reference model produced in Agisoft Metashape. Evaluation criteria included processing time, geometric accuracy (via M3C2 analysis), point cloud density and roughness, and point cloud completeness. Results show that radiance fields-based methods significantly reduce processing time while providing competitive visual results. SeaThru-NeRF demonstrated the highest geometric accuracy, benefiting from underwater-specific corrections, while 3DGS offered photorealistic rendering. These findings highlight the potential of neural methods for underwater cultural heritage documentation, though further improvements are needed in data fidelity and robustness under challenging underwater conditions.
@article{isprs-archives-XLVIII-M-9-2025-1475-2025, author = {Tanduo, B. and Matrone, F. and Murtiyoso, A.}, title = {Initial Experiments on the Use of Radiance Fields for Underwater 3D Reconstruction}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {XLVIII-M-9-2025}, year = {2025}, pages = {1475--1481}, doi = {10.5194/isprs-archives-XLVIII-M-9-2025-1475-2025}, dimensions = {true}, publisher = {Copernicus}, project = {GS3D}, }
ISPRS Archives
Implementing 3D Augmented Reality for Increasing Public Outreach of Majapahit Archeological Sites

F. Bioresita, H. Hidayat, A. Murtiyoso, and 5 more authors

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2025

Abs DOI Bib PDF

Cultural heritage sites, such as historical buildings and monuments, hold significant artistic, cultural, and historical value which necessitates preservation. This study explores the integration of 3D modelling and Augmented Reality (AR) technologies to digitally document and promote public engagement with heritage sites. Employing both laser scanning and photogrammetry, the research aims to develop accurate and photorealistic 3D models of key archaeological structures from the Majapahit era located in Trowulan, Indonesia—specifically Bajang Ratu, Wringin Lawang, and Brahu. Laser scanning demonstrates superior geometric accuracy and spatial completeness, producing denser point clouds and more detailed meshes, which are essential for precise documentation, restoration planning, and structural analysis. Photogrammetry, while offering lower geometric resolution, excels in capturing high-quality surface textures, making it more suitable for visual representation, public engagement, and AR applications. The findings highlight the benefits of a hybrid approach that combines laser scanning’s spatial precision with photogrammetry’s visual realism. This AR experience, when paired with a printed archaeological site map, allows users to interact with the digital reconstructions, enhancing historical understanding and accessibility. Furthermore, this research contributes to digital conservation efforts and innovative historical learning methods for Indonesian cultural heritage.
@article{isprs-archives-XLVIII-M-9-2025-141-2025, author = {Bioresita, F. and Hidayat, H. and Murtiyoso, A. and Moisan, E. and Grussenmeyer, P. and Cahyadi, M. N. and Cahyono, A. B. and Maulana, I.}, title = {Implementing 3D Augmented Reality for Increasing Public Outreach of Majapahit Archeological Sites}, journal = {The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {XLVIII-M-9-2025}, year = {2025}, pages = {141--147}, doi = {10.5194/isprs-archives-XLVIII-M-9-2025-141-2025}, dimensions = {true}, publisher = {Copernicus}, }

2024

ISPRS Archives
Comparison of state-of-the-art multi-view stereo solutions for close range heritage documentation

A. Murtiyoso, J.S. Markiewicz, A.K. Karwel, and 2 more authors

International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2024

Abs DOI Bib PDF

In recent years novel 3D reconstruction methods have been developed to improve the conventional image-based point cloud generation techniques. These novel methods generally attempt to address various challenges encountered in conventional methods, namely, the reconstruction of reflective surfaces and the amount of processing time required, both of which are major bottlenecks in heritage documentation and especially those related to large and complex objects. In this paper, we identified three types of 3D image-based reconstruction techniques and tested their usage on heritage datasets, namely (1) conventional multi-view stereo (MVS), (2) learning-based MVS, and (3) neural radiance fields (NeRF). The aim of this study is to determine the capabilities of these methods in reconstruction of three different heritage-related datasets with different challenges. Our results show that conventional MVS is nowadays a reliable solution for 3D reconstruction, in many instances recording good results relative to the reference terrestrial laser scans (TLS) when properly deployed. When applied to a challenging highly reflective scene, conventional MVS fared well using the PatchMatch algorithm (reaching an object completeness rate of 99.05%), while NeRF’s best performance was 99.98%. However, NeRF suffered from noisy data, some of which may stem from its radiance field-to-point cloud conversion method. The results show that there is great potential in using specific methods for specific cases, and research in combining them may yield interesting results in the future.
@article{murtiyoso2024comparison, title = {Comparison of state-of-the-art multi-view stereo solutions for close range heritage documentation}, author = {Murtiyoso, A. and Markiewicz, J.S. and Karwel, A.K. and Grussenmeyer, P. and Kot, P.}, journal = {International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences}, volume = {48}, number = {2}, pages = {317--323}, year = {2024}, doi = {10.5194/isprs-archives-XLVIII-2-W4-2024-317-2024}, dimensions = {true}, publisher = {Copernicus}, project = {GS3D}, }