Pith. sign in

REVIEW 3 major objections 8 minor 1 cited by

TUM2TWIN: Introducing the Large-Scale Multimodal Urban Digital Twin Benchmark Dataset

T0 review · 3 major / 8 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read TUM2TWIN claims to be the first comprehensive multimodal Urban Digital Twin benchmark, uniting 32 georeferenced subsets of point clouds, imagery, networks, and semantic 3D city models over about 100,000 square meters.

desk verdict A genuinely useful multimodal urban benchmark, but the accuracy table overreaches: LoD3 and LoD1 are assigned 2 cm absolute accuracy despite being derived from 0.5 m and 0.21 m source data, and the paper offers no validation protocol for either. read the letter →

arxiv 2505.07396 v2 pith:AZUFFKHL submitted 2025-05-12 cs.CV cs.LG

Olaf Wysocki , Benedikt Schwab , Manoj Kumar Biswanath , Michael Greza , Qilin Zhang , Jingwei Zhu , Thomas Froech , Medhini Heeramaglore
show 26 more authors
Ihab Hijazi Khaoula Kanna Mathias Pechinger Zhaiyu Chen Yao Sun Alejandro Rueda Segura Ziyang Xu Omar AbdelGafar Mansour Mehranfar Chandan Yeshwanth Yueh-Cheng Liu Hadi Yazdi Jiapan Wang Stefan Auer Katharina Anders Klaus Bogenberger Andre Borrmann Angela Dai Ludwig Hoegner Christoph Holst Thomas H. Kolbe Ferdinand Ludwig Matthias Nie{ss}ner Frank Petzold Xiao Xiang Zhu Boris Jutzi
This is my paper · ORCID
classification cs.CVcs.LG
keywords UrbanDigitalTwinbenchmarkdatasetmultimodaldatapointcloudsCityGMLLoD3buildingmodelsgeoreferencingsemanticsegmentation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that a single dataset can cover the full Urban Digital Twin processing chain, from raw acquisition to semantic 3D modeling, and introduces TUM2TWIN as that dataset. TUM2TWIN assembles 32 georeferenced subsets over roughly 100,000 square meters: terrestrial, mobile, drone, and aerial laser scans; street-level, thermal, drone, airplane, and satellite imagery; an HD road network; LoD1, LoD2, and LoD3 semantic building models; vegetation and CAD models; and a drone mesh. The result matters because most existing urban benchmarks stop at one stage or one modality, so reconstruction, segmentation, localization, and simulation methods cannot be checked against co-aligned ground truth from other sensors. The paper reports cm-level absolute accuracy for the TLS point cloud and the LoD3 models, and demonstrates the dataset on tasks including NeRF and Gaussian Splatting rendering, solar potential analysis, point cloud semantic segmentation, and LoD3 building reconstruction.

What carries the argument

The load-bearing object is the dataset itself, organized around georeferencing as the universal anchor: each subset is tied to a global coordinate reference system with a stated absolute and relative accuracy (Table 2), so any object in any modality can be located by the same x,y,z coordinates. Four pillars carry the structure — point clouds, images, networks, and 3D models — and a data dependency graph records how each subset was derived from source acquisitions. The georeferencing claim is what lets a point cloud, a photograph, a road centerline, and a semantic building model be treated as co-registered observations of the same scene rather than as separate benchmarks.

What would settle it

Take a set of independently surveyed ground control points across the campus and compare them against the coordinates of the TUM-TLS-24 scan and the LoD3 building models; if the residuals systematically exceed the stated 7 mm and 20 mm absolute accuracies by a substantial margin, the co-registration claim on which the benchmark's ground-truth value depends is falsified.

Watch

Extended reading notes

Core claim

TUM2TWIN is presented as the largest Urban Digital Twin benchmark dataset to date, with 32 data subsets, currently 767 GB, covering the same real urban area in a single global coordinate frame. Its central claim is that georeferencing is the key that lets all modalities — point clouds from terrestrial, mobile, drone, and airborne platforms; optical, thermal, and satellite imagery; a road network; semantic building models at LoD1–LoD3; vegetation models; CAD models; and a drone mesh — be overlaid and treated as mutual ground truth. On top of this, the paper reports a first-of-its-kind combination: LoD2, textured LoD2, and LoD3 models georeferenced together with TLS, UAS, and MLS point clouds, enabling ground-truth validation of LoD3 reconstruction from several sensors. The paper also surveys downstream work already using the dataset, including image-based vehicle localization, LoD1/LoD2/LoD3 reconstruction, facade segmentation and inpainting, thermal point cloud projection, NeRF and 3D Gaussian Splatting, driving simulators, and solar potential analysis.

Load-bearing premise

The load-bearing premise is that the centimeter-level georeferencing accuracies reported in Table 2 hold for every subset, so the point clouds, images, and 3D models genuinely occupy one common coordinate frame.

Editorial extensions

If this is right

  • A reconstruction or segmentation method can now be validated against co-aligned ground truth from another sensor, e.g., an MLS-based facade segmentation against the TLS point cloud and the LoD3 semantic model.
  • LoD3 building reconstruction can be benchmarked for the first time against high-accuracy TLS point clouds and manually modeled semantic LoD3 geometry in the same real-world scene.
  • Simulated-to-real transfer becomes directly testable because the simulated ALS subset and the real ALS point cloud cover the same buildings with the same LoD2 reference.
  • Multimodal registration can be studied where TUM-TLS-24 and TUM-MLS-24 overlap indoors and outdoors, with density and accuracy differences between the two mobile systems documented.
  • Novel view synthesis methods such as NeRF and 3D Gaussian Splatting can be scored against a high-accuracy TLS point cloud and low-poly semantic models rather than only against images or meshes.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the georeferencing holds at the stated accuracies, TUM2TWIN becomes a calibration anchor: any future sensor data tied to the same global frame — smartphone photogrammetry, new satellite imagery, other city scans — could be scored against the same ground truth without a new field campaign.
  • The multitemporal MLS and thermal sequences (2016, 2018, 2024) suggest a 4D change-detection benchmark, but the paper does not define the evaluation protocol; that task is left for the community.
  • The satellite layers (Sentinel-1 and Sentinel-2) contain very few pixels over the campus, so the interesting direction, which the paper raises but does not develop, is to use the cm-level 3D ground truth to label or interpret those coarse pixels.
  • Because the LoD3 models were manually built using the same MLS point clouds that serve as inputs to reconstruction evaluations, an independent audit of LoD3 against the TLS data would be needed before treating LoD3 as absolute ground truth; the paper does not provide that audit.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 8 minor

Summary. The paper introduces TUM2TWIN, a large-scale, georeferenced, multimodal urban digital twin dataset covering roughly 100,000 m^2 of the TUM main campus in Munich. It comprises 32 data subsets across four pillars: point clouds (TLS, MLS, UAS, ALS, and simulated ALS), images (street-level RGB and thermal, UAS orthophoto and raw imagery, airplane orthophoto, Sentinel-1/2, and simulated CuBy), networks (an HD map/road network), and 3D models (LoD1-3 semantic building models, streetspace, vegetation, tree, CAD, and mesh models). The paper describes the acquisition campaigns, reports per-subset absolute and relative accuracies in Table 2, and illustrates downstream use cases including vehicle localization, LoD1-3 reconstruction, facade segmentation, thermal projection, NeRF/3DGS, driving simulators, and solar potential analysis. The central claims are that TUM2TWIN is the first comprehensive multimodal urban digital twin benchmark dataset and the largest to date, with cm-level georeferencing accuracy enabling ground-truth validation of reconstruction and fusion methods.

Significance. If the accuracy and co-registration claims are upheld, TUM2TWIN would be a valuable community resource: a single testbed combining indoor-outdoor TLS, MLS, UAS, and ALS point clouds; street/thermal/aerial/satellite imagery; HD maps; and semantic LoD1-3 models, all in a global coordinate frame. The paper's strengths include the breadth of modalities, the clear data dependency graph (Figure 4), the acquisition timeline (Figure 3), the detailed TLS processing numbers (relative MAE 1.2 mm, absolute MAE 7.1 mm), and the fact that several companion papers (ZAHA, Scan2LoD3, Texture2LoD3) already use subsets of the data, providing early evidence of community uptake. However, the benchmark's core value as ground truth depends on the correctness of its accuracy figures and cross-modal alignment, and several of those figures are not substantiated by the described modeling workflows. These issues are addressable but must be resolved before the paper's central claims can be accepted at face value.

major comments (3)
  1. [Section 3.4.1 / Table 2] Section 3.4.1 states that LoD3 building models were created using '3D measurements of combined proprietary point clouds [56] and TUM-MLS-16,' yet Table 2 lists TUM-MLS-16 with 0.5 m absolute accuracy while crediting the LoD3 models with 0.02 m absolute accuracy. A model built from a 0.5 m source cannot inherit 2 cm absolute accuracy unless the proprietary MoSES data carries essentially all of the geometric burden, and that accuracy is not reported in the paper. The authors should either (i) provide a validation protocol comparing the LoD3 models against the TUM-TLS-24 point cloud or independent survey-grade checkpoints and report the resulting deviations, or (ii) reduce the claimed accuracy to the level supported by the stated source data. This is load-bearing because the benchmark's role as ground truth for LoD3 reconstruction, vehicle localization, and cross-modal fusion depends on the actual co-registration error.
  2. [Section 3.4.1 / Table 2] Section 3.4.1 says the LoD1 models' height information 'stemmed from the Real ALS,' which Table 2 lists with 0.21 m absolute accuracy, while footprints were extracted from LoD2 ground surfaces; nevertheless, LoD1 is assigned 0.02 m absolute accuracy in Table 2. This is internally inconsistent unless the footprint geometry fully determines the error budget, which is not stated. Please clarify how the 0.02 m figure was obtained, or correct the reported accuracy to a level consistent with the stated source data and error propagation.
  3. [Sections 4 and 5] The paper repeatedly calls TUM2TWIN a 'benchmark dataset,' but it does not define any standardized evaluation protocol: there are no task-specific metrics, no train/test splits, and no fixed evaluation subsets or leaderboard. The downstream sections (4.1-4.10) describe use cases and point to external companion papers (e.g., ZAHA) that do define benchmarks, but the present paper itself does not. To substantiate the 'benchmark' label, the authors should specify at least one concrete task with evaluation criteria and a fixed evaluation protocol, or reframe the contribution as a 'dataset' whose benchmarks are provided by the cited companion papers.
minor comments (8)
  1. [Section 4.6] The reference to 'Section 3.6.1' does not exist; it should be Section 3.4.1, which describes the LoD3 building models.
  2. [Abstract and Section 7] The claim that TUM2TWIN is 'the largest UDT benchmark dataset' is not quantified; please specify the metric (e.g., number of data subsets, spatial extent, data volume) and compare it with the datasets listed in Table 1.
  3. [Table 2] For Sentinel-1 and Sentinel-2, the values under 'Abs. Acc.' are actually spatial resolutions (5-40 m and 10-60 m), not absolute geolocation accuracy; please relabel the column or add a footnote clarifying that these are image resolutions.
  4. [Section 3.1.1] Please report the number of ground control points and the method used to establish their coordinates (e.g., total station, RTK-GNSS) for the reported 7.1 mm absolute georeferencing MAE of TUM-TLS-24.
  5. [Section 5.1] The text promises 'Binary openings' ground truth masks (Sec. 3.4.1),' but Section 3.4.1 does not describe such masks; either implement this data product or correct the reference.
  6. [Section 3.4.1] The 10 cm threshold for LoD3 facade elements (intrusion or extrusion) is a modeling parameter; please state how consistently it was applied across the dataset and whether it is reflected in the reported accuracy figures.
  7. [Table 1] The '# Data Instances' column compares different notions across datasets (scenes, objects, or data subsets); the table would be clearer if the unit were stated in the caption.
  8. [Data Availability] The dataset is only referenced via a project website; a persistent DOI and a clear license statement would improve the archival quality and usability of the benchmark.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation found; dataset claims rest on measured sensor specifications and external georeferencing, not on fitted outputs or self-citation chains.

full rationale

TUM2TWIN is a benchmark dataset contribution rather than a method paper; it makes no equation-level prediction whose output is defined by its inputs. The central claims (first comprehensive multimodal UDT benchmark, 32 subsets, cm-level georeferencing) are supported by acquisition specifications, external sources (LDBV, Sentinel missions, Pix4D, GCP/GNSS processing), and a comparison table, not by a self-referential derivation or a fitted parameter. The closest concern is the LoD3 accuracy entry in Table 2 (0.02 m) versus the modeling sources stated in Section 3.4.1, namely TUM-MLS-16 (0.5 m absolute accuracy in Table 2) and proprietary point clouds [56]. If the proprietary MoSES data does not carry the geometric burden, this is an unsupported or internally inconsistent quality claim, but it is not circular: the paper never derives LoD3 accuracy from the MLS accuracy, and no benchmarked method's output is defined in terms of the LoD3 model. Likewise, using manually modeled LoD3 models as ground truth for LoD3 reconstruction evaluated on the same MLS scans weakens independence, but this is a known benchmark-practice limitation rather than a reduction of the dataset's central claim to its own inputs. Self-citations such as Scan2LoD3, ZAHA, and Texture2LoD3 appear as example downstream tasks, not as load-bearing justification of the dataset's novelty, so they do not raise the circularity score.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

No fitted parameters or invented entities are central to the dataset's contribution. The single hand-chosen parameter is the 10 cm threshold for LoD3 facade detail. The key assumptions are about georeferencing accuracy and the reliability of third-party data sources.

free parameters (1)
  • LoD3 facade element threshold = 10 cm
    Section 3.4.1: 'if exceeding the threshold of intrusion or extrusion by 10 cm'. This threshold determines which facade elements (windows, doors, stairs) are modeled, so it shapes the LoD3 ground truth. It is a design choice, not independently derived.
assumptions (3)
  • domain assumption Georeferencing via a global CRS is sufficient to associate objects across modalities.
    Section 3 states that georeferencing 'serves as unique and earth-global positional identifier enabling association of any object in the dataset via globally defined xyz coordinates.' This assumes no systematic misregistration.
  • domain assumption The TLS registration and georeferencing errors (1.2 mm relative, 7.1 mm absolute) are correctly computed against B&W targets and GCPs.
    Section 3.1.1 reports these figures as the basis for the dataset's accuracy claims.
  • domain assumption The official Bavarian LoD2 building models have centimeter-level georeferencing accuracy.
    Section 3.4.1: 'The building models of the Bavarian State Mapping Agency cover the TUM main campus and are based on the official cadastre guaranteeing georeferencing accuracy in the centimeter range.'

how reviews work

0 comments
Cite this review

Pith. "Pith review of TUM2TWIN: Introducing the Large-Scale Multimodal Urban Digital Twin Benchmark Dataset." pith.science (2026). https://pith.science/paper/AZUFFKHL

@misc{pith2026250507396,
  author       = {Pith},
  title        = {Pith review of: TUM2TWIN: Introducing the Large-Scale Multimodal Urban Digital Twin Benchmark Dataset},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AZUFFKHL}},
  note         = {Machine review of arXiv:2505.07396}
}
abstract

Urban Digital Twins (UDTs) have become essential for managing cities and integrating complex, heterogeneous data from diverse sources. Creating UDTs involves challenges at multiple process stages, including acquiring accurate 3D source data, reconstructing high-fidelity 3D models, maintaining models' updates, and ensuring seamless interoperability to downstream tasks. Current datasets are usually limited to one part of the processing chain, hampering comprehensive UDTs validation. To address these challenges, we introduce the first comprehensive multimodal Urban Digital Twin benchmark dataset: TUM2TWIN. This dataset includes georeferenced, semantically aligned 3D models and networks along with various terrestrial, mobile, aerial, and satellite observations boasting 32 data subsets over roughly 100,000 $m^2$ and currently 767 GB of data. By ensuring georeferenced indoor-outdoor acquisition, high accuracy, and multimodal data integration, the benchmark supports robust analysis of sensors and the development of advanced reconstruction methods. Additionally, we explore downstream tasks demonstrating the potential of TUM2TWIN, including novel view synthesis of NeRF and Gaussian Splatting, solar potential analysis, point cloud semantic segmentation, and LoD3 building reconstruction. We are convinced this contribution lays a foundation for overcoming current limitations in UDT creation, fostering new research directions and practical solutions for smarter, data-driven urban environments. The project is available under: https://tum2t.win

Figures

Figures reproduced from arXiv: 2505.07396 by the authors.

Figure 1
Figure 1. TUM2TWIN: Georeferenced, semantic-rich, multimodal, multitemporal, and [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. TUM2TWIN covers approximately 100 000 m2 of the center of Munich, Germany (highlighted), boasting 32 data subsets and currently totaling 767 GB. 3 [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Timeline of the TUM2TWIN benchmark dataset. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (18 more)
Figure 4
Figure 4. Figure 4: Data dependency graph presenting the relation of the TUM2TWIN sub-datasets. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: TUM2TWIN data acquisition campaigns (w/o derived high-level representations [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Partial outdoor and indoor example scenarios included in TUM-TLS-24. The [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: A tree example from the TReeML-Data, presented in two different represen [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]
Figure 8
Figure 8. Figure 8: Colored point cloud of an indoor scene from TUM-MLS-24. [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]
Figure 9
Figure 9. Figure 9: To date, the largest facade semantic segmentation dataset: ZAHA. [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]
Figure 10
Figure 10. Figure 10: Overview of UAS data products of TUM main campus, including UAS Or [PITH_FULL_IMAGE:figures/full_fig_p016_10.png]
Figure 11
Figure 11. Figure 11: Example of TIR image of Street-Level Thermal Infrared 16. [PITH_FULL_IMAGE:figures/full_fig_p019_11.png]
Figure 12
Figure 12. Figure 12: Georeferenced Sentinel-1 images, quick looks for four quarters: a) Feb 09 2022, [PITH_FULL_IMAGE:figures/full_fig_p021_12.png]
Figure 13
Figure 13. Figure 13: Side-by-Side illustration of an aerial view image (left) and the digital twin [PITH_FULL_IMAGE:figures/full_fig_p024_13.png]
Figure 14
Figure 14. Figure 14: The primary difference between different LoDs: While LoD1 is an approxi [PITH_FULL_IMAGE:figures/full_fig_p026_14.png]
Figure 15
Figure 15. Figure 15: Volumetric geometry representations of traffic spaces with different heights [PITH_FULL_IMAGE:figures/full_fig_p027_15.png]
Figure 16
Figure 16. Figure 16: The TUM2TWIN-unlocked downstream tasks: Multiple research areas already [PITH_FULL_IMAGE:figures/full_fig_p029_16.png]
Figure 17
Figure 17. Figure 17: Exemplary facade image inpainting: a) Ground-Truth CM derived from se [PITH_FULL_IMAGE:figures/full_fig_p032_17.png]
Figure 18
Figure 18. Figure 18: Building model overlayed with 3D thermal point clouds for projection. [PITH_FULL_IMAGE:figures/full_fig_p033_18.png]
Figure 19
Figure 19. Figure 19: Comparison of reconstruction results using the TUM2TWIN dataset. (a) [PITH_FULL_IMAGE:figures/full_fig_p034_19.png]
Figure 20
Figure 20. Figure 20: Solar potential analysis (left) and photovoltaic panel placement (right) on the [PITH_FULL_IMAGE:figures/full_fig_p036_20.png]
Figure 21
Figure 21. Figure 21: A potential downstream tasks of multimodal registration of LoD2 models and [PITH_FULL_IMAGE:figures/full_fig_p039_21.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. To Glue or Not to Glue? Classical vs Learned Image Matching for Mobile Mapping Cameras to Textured Semantic 3D Building Models

    cs.CV 2025-05 conditional novelty 4.0 of 10

    Learned feature matchers, especially SuperPoint plus LightGlue, outperform handcrafted SIFT, ORB, and AKAZE when matching mobile mapping images to textured CityGML LoD2 building models for absolute pose estimation.

Reference graph

Works this paper leans on

86 extracted references · 64 canonical work pages · cited by 1 Pith paper

  1. [56]

    Accessed: 2023-01- 30

    3D Mapping Solutions, MoSES mobile mapping platform - technical details,https://www.3d-mapping.de/ueber-uns/unternehmensbereiche/ data-acquisition/unser-vermessungssystem/, 2023. Accessed: 2023-01- 30

    work page 2023

  2. [1]

    B. D. Allen, Digital twins and living models at NASA, in: Digital Twin Summit, 2021

    2021

  3. [2]

    doi:https://doi.org/10.1016/j

    T.Deng, K.Zhang, Z.-J.M.Shen, Asystematicreviewofadigitaltwincity: A new pattern of urban governance toward smart cities, Journal of Management Science and Engineering 6 (2021) 125–134. doi:https://doi.org/10.1016/j. jmse.2021.03.003

    doi:10.1016/j 2021

  4. [3]

    T. H. Kolbe, A. Donaubauer, Semantic 3D city modeling and BIM, in: W. Shi, M. F. Goodchild, M. Batty, M.-P. Kwan, A. Zhang (Eds.), Urban Informatics, Springer Singapore, Singapore, 2021, pp. 609–636

    2021

  5. [4]

    Haala, M

    N. Haala, M. Kada, An update on automatic 3D building reconstruction, ISPRS Journal of Photogrammetry and Remote Sensing 65 (2010) 570 – 580

    2010

  6. [5]

    Grieves, Digital twin: manufacturing excellence through virtual factory replication, White paper 1 (2014) 1–7

    M. Grieves, Digital twin: manufacturing excellence through virtual factory replication, White paper 1 (2014) 1–7

    2014

  7. [6]

    Wysocki, Y

    O. Wysocki, Y. Xia, M. Wysocki, E. Grilli, L. Hoegner, D. Cremers, U. Stilla, Scan2LoD3: Reconstructing semantic 3D building models at LoD3 using ray casting and Bayesian networks, IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2023) 6547–6557

    2023

  8. [7]

    Scharstein, R

    D. Scharstein, R. Szeliski, A taxonomy and evaluation of dense two-frame stereo correspondence algorithms, International journal of computer vision 47 (2002) 7–42. 41

    2002

Show all 86 references
  1. [8]

    Kirillov, E

    A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y. Lo, et al., Segment anything, in: Proceed- ings of the IEEE/CVF international conference on computer vision, 2023, pp. 4015–4026

    2023

  2. [9]

    Hackel, N

    T. Hackel, N. Savinov, L. Ladicky, J. D. Wegner, K. Schindler, M. Pollefeys, Semantic3d.net: Anewlarge-scalepointcloudclassificationbenchmark, arXiv preprint arXiv:1704.03847 (2017)

    2017 arXiv

  3. [10]

    De Deuge, A

    M. De Deuge, A. Quadros, C. Hung, B. Douillard, Unsupervised feature learning for classification of outdoor 3d scans, Australasian Conference on Robotics and Automation 2 (2013)

    2013

  4. [11]

    Matrone, A

    F. Matrone, A. Lingua, R. Pierdicca, E. S. Malinverni, M. Paolanti, E. Grilli, F. Remondino, A. Murtiyoso, T. Landes, A benchmark for large-scale her- itage point cloud semantic segmentation, The International Archives of the Photogrammetry, Remote Sensing and Spatial Informat...

    2020

  5. [12]

    Deschaud, D

    J.-E. Deschaud, D. Duque, J. P. Richa, S. Velasco-Forero, B. Marcotegui, F. Goulette, Paris-CARLA-3D: A real and synthetic outdoor point cloud dataset for challenging tasks in 3D mapping, Remote Sensing 13 (2021) 4713

    2021

  6. [13]

    W. Tan, N. Qin, L. Ma, Y. Li, J. Du, G. Cai, K. Yang, J. Li, Toronto- 3D: A large-scale mobile LiDAR dataset for semantic segmentation of urban roadways, in: IEEE/CVF Conference on Computer Vision and Pattern Recog- nition Workshops (CVPRW), 2020, pp. 202–203

    2020

  7. [14]

    Serna, B

    A. Serna, B. Marcotegui, F. Goulette, J.-E. Deschaud, Paris-rue-Madame database: As 3D mobile laser scanner dataset for benchmarking urban detec- tion, segmentation and classification methods, in: Proceedings of the Interna- tional Conference on Pattern Recognition Application...

    2014

  8. [15]

    Stilla, Y

    U. Stilla, Y. Xu, Change detection of urban objects using 3d point clouds: A review, ISPRS Journal of Photogrammetry and Remote Sensing 197 (2023) 228–255. doi:https://doi.org/10.1016/j.isprsjprs.2023.01.010

    2023 doi

  9. [16]

    Armeni, O

    I. Armeni, O. Sener, A. R. Zamir, H. Jiang, I. Brilakis, M. Fischer, S. Savarese, 3D semantic parsing of large-scale indoor spaces, in: Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 1534–1543. 42

    2016

  10. [17]

    F. Nex, E. Stathopoulou, F. Remondino, M. Yang, L. Madhuanand, Y. Yo- gender, B. Alsadik, M. Weinmann, B. Jutzi, R. Qin, Usegeo-a uav-based multi-sensor dataset for geospatial research, ISPRS Open Journal of Pho- togrammetry and Remote Sensing (2024) 100070

    2024

  11. [18]

    Z. Dong, F. Liang, B. Yang, Y. Xu, Y. Zang, J. Li, Y. Wang, W. Dai, H. Fan, J. Hyyppä, et al., Registration of large-scale terrestrial laser scanner point clouds: A review and benchmark, ISPRS Journal of Photogrammetry and Remote Sensing 163 (2020) 327–342

    2020

  12. [19]

    Schwab, S

    B. Schwab, S. Haas Goschenhofer, O. Wysocki, LoD3 road space models, release v0.8.1,https://github.com/savenow/lod3-road-space-models,

  13. [20]

    Kölle, D

    M. Kölle, D. Laupheimer, S. Schmohl, N. Haala, F. Rottensteiner, J. D. Weg- ner, H. Ledoux, The hessigheim 3d (h3d) benchmark on semantic segmenta- tion of high-resolution 3d point clouds and textured meshes from uav lidar and multi-view-stereo, ISPRSOpenJournalofPhotogrammetr...

    2021

  14. [21]

    W. Gao, L. Nan, B. Boom, H. Ledoux, Sum: A benchmark dataset of semantic urban meshes, ISPRS Journal of Photogrammetry and Remote Sensing 179 (2021) 108–120. doi:https://doi.org/10.1016/j.isprsjprs.2021.07.008

    2021 doi

  15. [22]

    Y. Liao, J. Xie, A. Geiger, KITTI-360: A novel dataset and benchmarks for urban scene understanding in 2D and 3D, arXiv preprint arXiv:2109.13410 (2021)

    2021 arXiv

  16. [23]

    L. Lin, Y. Liu, Y. Hu, X. Yan, K. Xie, H. Huang, Capturing, reconstruct- ing, and simulating: the urbanscene3d dataset, in: European Conference on Computer Vision, Springer, 2022, pp. 93–109

    2022

  17. [24]

    R. Wang, S. Huang, H. Yang, Building3D: A urban-scale dataset and bench- marks for learning roof structures from point clouds, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 20076– 20086

    2023

  18. [25]

    S. M. González-Collazo, J. Balado, I. Garrido, J. Grandío, R. Rashdi, E. Tsir- anidou, P. del Río-Barral, E. Rúa, I. Puente, H. Lorenzo, Santiago urban dataset sud: Combination of handheld and mobile laser scanning point clouds, Expert Systems with Applications 238 (2024) 121842. 43

    2024

  19. [26]

    W. Gao, R. Peters, J. Stoter, Building-pcc: Building point cloud comple- tion benchmarks, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences X-4/W5-2024 (2024) 179–186. doi:10.5194/ isprs-annals-X-4-W5-2024-179-2024

    2024

  20. [27]

    Yazdi, Q

    H. Yazdi, Q. Shu, T. Rötzer, F. Petzold, F. Ludwig, A multilayered urban tree dataset of point clouds, quantitative structure and graph models, Scientific Data 11 (2024) 28

    2024

  21. [28]

    Raumonen, M

    P. Raumonen, M. Kaasalainen, M. Åkerblom, S. Kaasalainen, H. Kaartinen, M. Vastaranta, M. Holopainen, M. Disney, P. Lewis, Fast automatic precision tree models from terrestrial laser scanner data, Remote Sensing 5 (2013) 491–520

    2013

  22. [29]

    Yazdi, Q

    H. Yazdi, Q. Shu, T. Rötzer, F. Petzold, F. Ludwig, Treeml-data; a multidis- ciplinary and multilayer urban tree dataset, Springer Nature (2023)

    2023

  23. [30]

    J. Zhu, J. Gehrung, R. Huang, B. Borgmann, Z. Sun, L. Hoegner, M. Hebel, Y. Xu, U. Stilla, TUM-MLS-2016: An annotated mobile LiDAR dataset of the TUM City Campus for semantic point cloud interpretation in urban areas, Remote Sensing 12 (2020) 1875

    2020

  24. [31]

    Wysocki, Y

    O. Wysocki, Y. Tan, T. Froech, Y. Xia, M. Wysocki, L. Hoegner, D. Cremers, C. Holst, Zaha: Introducing the level of facade generalization and the large- scale point cloud facade semantic segmentation benchmark dataset, 2024. URL:https://arxiv.org/abs/2411.04865.arXiv:2411.04865

    2024 arXiv

  25. [32]

    Wysocki, L

    O. Wysocki, L. Hoegner, U. Stilla, TUM-FAÇADE: Reviewing and enriching point cloud benchmarks for façade segmentation, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLVI-2/W1-2022 (2022) 529–536

    2022

  26. [33]

    Anders, J

    K. Anders, J. Wang, M. Chang, M. Letard, F. Schulte, L. Winiwarter, Terrestrial and UAV laser scanning point clouds of TUM Campus Otto- brunn, 2024. URL:https://doi.org/10.5281/zenodo.14443336. doi:10. 5281/zenodo.14443336

    2024 doi

  27. [34]

    Pix4D SA, Pix4Dmatic Software, Version 1.71.0,https://www.pix4d.com/ product/pix4dmatic(24 April 2025), 2024

    2025

  28. [35]

    J. L. Schönberger, E. Zheng, M. Pollefeys, J.-M. Frahm, Pixelwise View Selection for Unstructured Multi-View Stereo, in: European Conference on Computer Vision (ECCV), 2016. 44

    2016

  29. [36]

    Accessed: 2025-03-22

    LDBV, Das Landesamt für Digitalisierung, Breitband und Ver- messung (LDBV),https://www.ldbv.bayern.de/produkte/ landschaftsinformationen/laser.html, 2025. Accessed: 2025-03-22

    2025

  30. [37]

    Winiwarter, A

    L. Winiwarter, A. M. E. Pena, H. Weiser, K. Anders, J. M. Sánchez, M. Searle, B. Höfle, Virtual laser scanning with helios++: A novel take on ray tracing- based simulation of topographic full-waveform 3d laser scanning, Remote Sensing of Environment 269 (2022) 112772

    2022

  31. [38]

    Accessed: 2025-03-22

    Google, Google Street View,https://www.google.com/maps, 2023. Accessed: 2025-03-22

    2023

  32. [39]

    W. Tang, W. Li, X. Liang, O. Wysocki, F. Biljecki, C. Holst, B. Jutzi, Tex- ture2LoD3: Enabling LoD3 Building Reconstruction With Panoramic Images, https://arxiv.org/abs/2504.05249 (2025).arXiv:2504.05249

    2025 arXiv

  33. [40]

    J. Zhu, Y. Xu, L. Hoegner, U. Stilla, Generation of thermal point clouds from uncalibrated thermal infrared image sequences and mobile laser scans, IEEE Transactions on Instrumentation and Measurement 72 (2023) 1–16. doi:10. 1109/TIM.2023.3284942

    2023

  34. [41]

    Accessed: 2025-03-22

    LDBV, Das Landesamt für Digitalisierung, Breitband und Vermes- sung (LDBV),https://www.ldbv.bayern.de/produkte/luftbilder/ orthophotos/bayernbefliegung.html, 2025. Accessed: 2025-03-22

    2025

  35. [42]

    Torres, I

    R. Torres, I. Navas-Traver, D. Bibby, S. Lokas, P. Snoeij, B. Rommen, S. Os- borne, F. Ceba-Vega, P. Potin, D. Geudtner, Sentinel-1 SAR system and mission, in: 2017 IEEE Radar Conference (RadarConf), 2017, pp. 1582–1585. doi:10.1109/RADAR.2017.7944460

    2017

  36. [43]

    Spoto, O

    F. Spoto, O. Sy, P. Laberinti, P. Martimort, V. Fernandez, O. Colin, B. Hoersch, A. Meygret, Overview of Sentinel-2, in: 2012 IEEE Inter- national Geoscience and Remote Sensing Symposium, 2012, pp. 1707–1710. doi:10.1109/IGARSS.2012.6351195

    2012

  37. [44]

    Espesset, B

    C.Bouzinac, B.Lafrance, L.Pessiot, D.Touli, M.Jung, S.Massera, M.Neveu- VanMalle, A. Espesset, B. Francesconi, S. Clerc, J. Jackson, B. Alhammoud, F. Viallefont, E. Cadau, R. Iannone, F. Gascon, Sentinel-2 level-1 calibration and validation status from the mission performance ...

    2018

  38. [45]

    Hagolle, M

    O. Hagolle, M. Huc, C. Desjardins, S. Auer, R. Richter, MAJA algorithm theoretical basis document, 2017. URL:https://doi.org/10.5281/zenodo. 1209633. doi:10.5281/zenodo.1209633

    2017 doi

  39. [46]

    Greza, L

    M. Greza, L. Hoegner, P.-R. Hirt, R. Roschlaub, U. Stilla, Satel- lite network bavaria - mission and data processing, Publikationen der Deutschen Gesellschaft für Photogrammetrie, Fernerkundung und Geoinfor- mation (DGPF) 43 (2023) 174–182

    2023

  40. [47]

    N. Lenz, M. Greza, Simulation of earth observation data utilizing a virtual satellite camera, Publikationen der Deutschen Gesellschaft für Photogramme- trie, Fernerkundung und Geoinformation (DGPF) 43 (2023) 50–60

    2023

  41. [48]

    Wysocki, B

    O. Wysocki, B. Schwab, C. Beil, C. Holst, T. H. Kolbe, Reviewing open data semantic 3D city models to develop novel 3D reconstruction methods, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 48 (2024) 493–500

    2024

  42. [49]

    T. H. Kolbe, T. Kutzner, C. S. Smyth, C. Nagel, C. Roensdorf, C. Heazel, OGC City Geography Markup Language (CityGML) Part 1: Conceptual Model Standard v3.0, 2021

    2021

  43. [50]

    Kutzner, C

    T. Kutzner, C. Smyth, C. Nagel, V. Coors, D. Vinasco-Alvarez, N. Ishimaru, Z. Yao, C. Heazel, T. H. Kolbe, OGC City Geography Markup Language (CityGML) Version 3.0 Part 2: GML Encoding Standard, Open Geospatial Consortium, 2023

    2023

  44. [51]

    Biljecki, J

    F. Biljecki, J. Stoter, H. Ledoux, S. Zlatanova, A. Çöltekin, Applications of 3D city models: State of the art review, ISPRS International Journal of Geo-Information 4 (2015) 2842–2889

    2015

  45. [52]

    Z. Yao, C. Nagel, F. Kunde, G. Hudra, P. Willkomm, A. Donaubauer, T. Adolphi, T. H. Kolbe, 3DCityDB - a 3D geodatabase solution for the management, analysis, and visualization of semantic 3D city models based on CityGML, Open Geospatial Data, Software and Standards 3 (2018). d...

    2018 doi

  46. [53]

    Ledoux, F

    H. Ledoux, F. Biljecki, B. Dukai, K. Kumar, R. Peters, J. Stoter, T. Comman- deur, 3dfier: Automatic reconstruction of 3d city models, Journal of Open Source Software 6 (2021) 2866. doi:10.21105/joss.02866. 46

    2021 doi

  47. [54]

    Huang, 3D Cities: Evaluating 3D building reconstruction tools using point clouds, Master’s thesis, Department Aerospace and Geodesy, Technical Uni- versity of Munich, 2024

    C. Huang, 3D Cities: Evaluating 3D building reconstruction tools using point clouds, Master’s thesis, Department Aerospace and Geodesy, Technical Uni- versity of Munich, 2024

    2024

  48. [55]

    Roschlaub, J

    R. Roschlaub, J. Batscheider, An INSPIRE-conform 3D building model of Bavaria using cadastre information, LiDAR and image matching, The Inter- national Archives of the Photogrammetry, Remote Sensing and Spatial Infor- mation Sciences XLI-B4 (2016) 747–754

    2016

  49. [57]

    C. Beil, T. H. Kolbe, Applications for semantic 3d streetspace models and their requirements—a review and look at the road ahead, ISPRS International Journal of Geo-Information 13 (2024). doi:10.3390/ijgi13100363

    2024 doi

  50. [58]

    Schwab, C

    B. Schwab, C. Beil, T. H. Kolbe, Spatio-semantic road space modeling for ve- hicle–pedestrian simulation to test automated driving systems, Sustainability 12 (2020). doi:10.3390/su12093799

    2020 doi

  51. [59]

    Münzinger, N

    M. Münzinger, N. Prechtel, M. Behnisch, Mapping the urban forest in detail: From lidar point clouds to 3d tree models, Urban Forestry & Urban Greening 74 (2022) 127637. doi:https://doi.org/10.1016/j.ufug.2022.127637

    2022

  52. [60]

    S. Zagst, Automatisierte Generierung eines Baumkatasters aus Punktwolken in unterschiedlichen urbanen Umgebungen, Master’s thesis, Department Aerospace and Geodesy, Technical University of Munich, 2023

    2023

  53. [61]

    Bieringer, O

    A. Bieringer, O. Wysocki, S. Tuttas, L. Hoegner, C. Holst, Analyzing the impact of semantic lod3 building models on image-based vehicle localization, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Informa- tion Sciences 10 (2024) 55–62

    2024

  54. [62]

    Z. Chen, Y. Shi, L. Nan, Z. Xiong, X. X. Zhu, Polygnn: Polyhedron-based graph neural network for 3d building reconstruction from point clouds, IS- PRS Journal of Photogrammetry and Remote Sensing 218 (2024) 693–706. doi:https://doi.org/10.1016/j.isprsjprs.2024.09.031

    2024 doi

  55. [63]

    Wysocki, L

    O. Wysocki, L. Hoegner, U. Stilla, MLS2LoD3: Refining low lods building models with mls point clouds to reconstruct semantic lod3 building models, 47 International 3D GeoInfo Conference 2023, Recent Advances in 3D Geoinfor- mation Science (2023) 367–380

    2023

  56. [64]

    Y. Wang, W. Jiao, H. Fan, G. Zhou, A framework for fully automated recon- struction of semantic building model at urban-scale using textured lod2 data, ISPRS Journal of Photogrammetry and Remote Sensing 216 (2024) 90–108

    2024

  57. [65]

    Weinmann, B

    M. Weinmann, B. Jutzi, C. Mallet, Feature relevance assessment for the se- manticinterpretationof3Dpointclouddata, ISPRSAnnalsofthePhotogram- metry, Remote Sensing and Spatial Information Sciences II-5/W2 (2013) 313– 318

    2013

  58. [66]

    Y. Tan, O. Wysocki, L. Hoegner, U. Stilla, Classifying point clouds at the facade-level using geometric features and deep learning networks, Accepted to proceedings of 3D GeoInfo 2023, Lecture Notes in Geoinformation and Cartography (2023)

    2023

  59. [67]

    Wysocki, E

    O. Wysocki, E. Grilli, L. Hoegner, U. Stilla, Combining visibility analysis and deep learning for refinement of semantic 3D building models by conflict classification, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences X-4/W2-2022 (2022) 289–296

    2022

  60. [68]

    S.Tuttas, U.Stilla, A.Braun, A.Borrmann, ValidationofBIMcomponentsby photogrammetric point clouds for construction site monitoring, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences II- 3/W4 (2015) 231–237

    2015

  61. [69]

    Fröch, O

    T. Fröch, O. Wysocki, Y. Xia, J. Xie, B. Schwab, D. Cremers, T. H. Kolbe, Facadiffy: Inpainting unseen facade parts using diffusion models, Accepted for ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Infor- mation Sciences (ISPRS Geospatial Week) (2025)

    2025

  62. [70]

    Tuttas, U

    S. Tuttas, U. Stilla, Reconstruction of façades in point clouds from multi aspect oblique ALS, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences II-3/W3 (2013) 91–96

    2013

  63. [71]

    J. Zhu, Y. Xu, Z. Ye, L. Hoegner, U. Stilla, Fusion of urban 3d point clouds with thermal attributes using MLS data and TIR image sequences, Infrared Physics & Technology 113 (2021) 103622. doi:10.1016/j.infrared. 2020.103622

    2021

  64. [72]

    M. K. Biswanath, L. Hoegner, U. Stilla, Thermal mapping from point clouds to 3d building model facades, Remote Sensing 15 (2023) 4830. 48

    2023

  65. [73]

    Borrmann, M

    A. Borrmann, M. K. Biswanath, A. Braun, C. Zhaiyu, D. Cremers, M. Heera- maglore, L. Hoegner, M. Mehranfar, T. H. Kolbe, F. Petzold, A. Rueda, S. Solonets, X. Zhu, Artificial intelligence for the automated creation of multi- scale digital twins of the built world – ai4twinning...

    2023

  66. [74]

    Mildenhall, P

    B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi, R. Ng, Nerf: Representing scenes as neural radiance fields for view synthesis, Communications of the ACM 65 (2021) 99–106

    2021

  67. [75]

    Petrovska, B

    I. Petrovska, B. Jutzi, Vision through obstacles—3D geometric reconstruction and evaluation of Neural Radiance Fields (NeRFs), Remote Sensing 16 (2024) 1188

    2024

  68. [76]

    Salahi, A

    M.Tancik, E.Weber, E.Ng, R.Li, B.Yi, T.Wang, A.Kristoffersen, J.Austin, K. Salahi, A. Ahuja, et al., Nerfstudio: A modular framework for neural radi- ance field development, in: ACM SIGGRAPH 2023 Conference Proceedings, 2023, pp. 1–12

    2023

  69. [77]

    J. T. Barron, B. Mildenhall, D. Verbin, P. P. Srinivasan, P. Hedman, Mip- nerf 360: Unbounded anti-aliased neural radiance fields, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 5470–5479

    2022

  70. [78]

    Müller, A

    T. Müller, A. Evans, C. Schied, A. Keller, Instant neural graphics primitives with a multiresolution hash encoding, ACM transactions on graphics (TOG) 41 (2022) 1–15

    2022

  71. [79]

    Kerbl, G

    B. Kerbl, G. Kopanas, T. Leimkühler, G. Drettakis, 3d gaussian splatting for real-time radiance field rendering., ACM Trans. Graph. 42 (2023) 139–1

    2023

  72. [80]

    Zhang, O

    Q. Zhang, O. Wysocki, S. Urban, B. Jutzi, Cdgs: Confidence-aware depth regularization for 3d gaussian splatting, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 48 (2024) 189–196

    2024

  73. [81]

    Willenborg, M

    B. Willenborg, M. Pültz, T. H. Kolbe, Integration of semantic 3d city models and 3d mesh models for accuracy improvements of solar potential analyses, The International Archives of the Photogrammetry, Remote Sensing and Spa- tial Information Sciences 42 (2018) 223–230

    2018

  74. [82]

    Rueda Segura, Y

    A. Rueda Segura, Y. Sun, I. Bratoev, F. Petzold, Generating façade seg- mentation datasets using diffusion models, in: Computer-Aided Architectural 49 Design Research in Asia (CAADRIA), 2025. Accepted for publication, forth- coming

    2025

  75. [83]

    Heeramaglore, T

    M. Heeramaglore, T. H. Kolbe, Semantically enriched voxels as a common representation for comparison and evaluation of 3d building models, in: Proceedings of the 17th International 3D GeoInfo Con- ference 2022, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Inf...

    2022 doi

  76. [84]

    Yeshwanth, Y.-C

    C. Yeshwanth, Y.-C. Liu, M. Nießner, A. Dai, ScanNet++: A high-fidelity dataset of 3d indoor scenes, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 12–22

    2023

  77. [85]

    Dubois, B

    C. Dubois, B. Jutzi, M. Olijslagers, C. Pathe, C. Schmullius, M. A. Stelmaszczuk-Górska, D.Vandenbroucke, M.Weinmann, Knowledgeandskills related to active optical sensors in the body of knowledge for earth observation and geoinformation (EO4GEO BOK), ISPRS Annals of the Photog...

    2021 doi

  78. [2021]

    Accessed: 2023-01-30

    2023

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.