pith. sign in

arxiv: 2606.19154 · v1 · pith:V4WB7EAJnew · submitted 2026-06-17 · 💻 cs.RO

Viking Hill Dataset: A Lidar-Radar-Camera Dataset for Detection and Segmentation in Forest Scenes

Vladim\'ir Kubelka , Oleksandr Kotlyar , Unal Artan , Martin Magnusson This is my paper

Pith reviewed 2026-06-26 20:42 UTC · model grok-4.3

classification 💻 cs.RO
keywords forest datasetimaging radarlidarsemantic segmentationautonomous navigationsensor fusiontree detectionpoint cloud
0
0 comments X

The pith

A new multi-sensor forest dataset shows imaging radar achieving IoU scores competitive with lidar for ground and canopy segmentation.

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

The paper introduces a dataset of forest scenes recorded by a mobile robot carrying a high-resolution FMCW imaging radar, lidar, RGB camera, IMU and RTK-GNSS. Data were collected across two sessions with different vegetation conditions and supplied with 3D cuboid annotations that assign consistent semantic labels to all three perception modalities. Baseline MinkowskiUNet segmentation on the point clouds yields radar IoU values of 91 percent for ground and 86 percent for canopy, close to lidar results, while radar trails on finer structures such as trunks. The dataset is released publicly to support work on mapping, localization and sensor fusion where visual and lidar signals can be blocked or degraded by canopy and weather.

Core claim

The Viking Hill Dataset supplies co-registered high-resolution FMCW radar, lidar and camera recordings from forest sites under contrasting vegetation states together with 3D cuboid annotations that include per-tree diameter estimates. These shared labels enable direct comparison of semantic segmentation performance across modalities, revealing that radar reaches IoU scores of 91 percent on ground and 86 percent on canopy, competitive with lidar, while lagging on tree trunks at 56 percent versus lidar's 74 percent.

What carries the argument

The 3D cuboid annotations that assign shared semantic labels across radar, lidar and camera point clouds, enabling modality-comparative segmentation benchmarks.

If this is right

  • Radar can serve as a reliable primary sensor for large-scale classes such as ground and canopy in forest navigation systems.
  • Diameter-stratified evaluation shows segmentation quality improves with larger tree size for both modalities.
  • Cross-modality comparison against RGB detection models highlights where radar complements camera-based trunk detection.
  • The co-registered data and precise RTK-GNSS positioning enable development of multi-sensor mapping and localization pipelines.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The dataset could be used to train fusion networks that retain radar's resilience while adding lidar detail on trunks.
  • Seasonal contrast between the two sessions provides a ready testbed for assessing long-term robustness of perception algorithms.
  • Future extensions might add dynamic elements or additional radar operating modes to probe occlusion handling.

Load-bearing premise

The 3D cuboid annotations give accurate shared semantic labels across the three sensors and the two recording sessions represent typical contrasting forest conditions.

What would settle it

Re-running MinkowskiUNet on the released point clouds and finding radar IoU for ground or canopy more than ten points below the reported lidar figures would falsify the competitiveness claim.

Figures

Figures reproduced from arXiv: 2606.19154 by Martin Magnusson, Oleksandr Kotlyar, Unal Artan, Vladim\'ir Kubelka.

Figure 1
Figure 1. Figure 1: Three modalities provided in the dataset: RGB images, lidar, and [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The dataset area, mapped by a mobile robot equipped with an [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Sensor suite on the Clearpath Husky robot. [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Top-down view on the two recording sessions; [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: A segment of the reference lidar map from the low vegetation [PITH_FULL_IMAGE:figures/full_fig_p013_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The training-validation-testing data split for the semantic seg [PITH_FULL_IMAGE:figures/full_fig_p016_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Visualization of the inference results on the radar testing set. For [PITH_FULL_IMAGE:figures/full_fig_p017_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Visualization of steps in the tree-trunk detection cross-modality [PITH_FULL_IMAGE:figures/full_fig_p021_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: RGB image, lidar and radar detection recall evaluated in five [PITH_FULL_IMAGE:figures/full_fig_p022_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Distribution of annotated trees visible per lidar scan as a function [PITH_FULL_IMAGE:figures/full_fig_p024_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Examples of annotation fail cases. (top) Tree trunk annotation [PITH_FULL_IMAGE:figures/full_fig_p026_11.png] view at source ↗
read the original abstract

Autonomous robots operating under forest canopies need robust perception of trees and surrounding vegetation across varying seasonal conditions. Existing forestry datasets provide lidar or camera data with per-tree annotations, but none include co-registered 4D imaging radar -- a modality of growing interest for its resilience to visual degradation, surface contamination, and vegetation occlusion. We introduce a multi-sensor forest dataset collected by a mobile robot equipped with a high-resolution FMCW imaging radar, lidar, RGB camera, IMU, and RTK-GNSS. The site was recorded in two sessions under contrasting vegetation states, and 3D cuboid annotations -- including per-tree diameter estimates -- provide shared semantic labels across all three perception modalities. Furthermore, we provide baseline results for semantic segmentation of the radar and lidar point clouds using MinkowskiUNet. Radar achieves IoU scores competitive with lidar for dominant classes (ground 91%, canopy 86%) while lagging on geometrically fine structures such as tree trunks (56% vs. 74%). A cross-modality analysis further compares lidar and radar trunk segmentation against an RGB detection model, and a diameter-stratified evaluation reveals how trunk segmentation quality varies with tree size. Beyond segmentation, the co-registered multi-modal data and RTK-GNSS-aided reference positioning support research in mapping, localization, and sensor fusion under canopy. The dataset and annotation tools are publicly available.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The paper introduces the Viking Hill Dataset, a multi-sensor collection (high-resolution FMCW imaging radar, lidar, RGB camera, IMU, RTK-GNSS) recorded by a mobile robot over two sessions with contrasting vegetation states in a forest. It supplies 3D cuboid annotations (including per-tree diameters) that serve as shared semantic labels across modalities and reports baseline semantic segmentation results on radar and lidar point clouds using MinkowskiUNet. Radar is stated to achieve IoU scores competitive with lidar on dominant classes (ground 91%, canopy 86%) but lower on tree trunks (56% vs. 74%), with additional cross-modality trunk analysis against an RGB detector and diameter-stratified evaluation. The dataset, annotations, and tools are released publicly to support mapping, localization, and fusion research under canopy.

Significance. If the shared 3D cuboid labels prove reliable, the release supplies a needed multi-modal forest benchmark that includes 4D imaging radar, a modality whose occlusion and weather resilience is of growing interest. The two-session design, diameter estimates, and public baselines enable direct comparison of sensor performance on the same scenes and support downstream work on sensor fusion and robust perception.

major comments (1)
  1. [Abstract] Abstract: the headline IoU comparisons (radar ground 91%/canopy 86%/trunk 56% vs. lidar trunk 74%) and the claim of radar competitiveness rest on the assumption that the 3D cuboid annotations supply accurate, modality-agnostic semantic labels. No annotation protocol, inter-annotator agreement statistics, or validation against independent measurements (e.g., manual diameter checks or lidar-only vs. camera-only consistency) is described; if cuboids were drawn primarily from one modality and projected, label noise on fine structures could systematically favor lidar and invalidate the cross-modality analysis.
minor comments (1)
  1. [Abstract] Abstract: the baseline method is named 'MinkowskiUNet' without citation, training details, or hyper-parameter settings, making the reported IoU numbers difficult to reproduce from the given information alone.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive review and the recognition of the dataset's potential value for multi-modal forest perception research. The major comment correctly identifies that the manuscript does not describe the annotation protocol, inter-annotator statistics, or independent validation of the 3D cuboids. We address this point below and will revise the paper accordingly.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the headline IoU comparisons (radar ground 91%/canopy 86%/trunk 56% vs. lidar trunk 74%) and the claim of radar competitiveness rest on the assumption that the 3D cuboid annotations supply accurate, modality-agnostic semantic labels. No annotation protocol, inter-annotator agreement statistics, or validation against independent measurements (e.g., manual diameter checks or lidar-only vs. camera-only consistency) is described; if cuboids were drawn primarily from one modality and projected, label noise on fine structures could systematically favor lidar and invalidate the cross-modality analysis.

    Authors: We agree that the absence of an explicit annotation protocol description weakens the cross-modality claims. The cuboids were generated by projecting 3D points from the lidar scans into the camera images for visual verification and then fitting cuboids with diameter estimates taken directly from the lidar data; however, no formal inter-annotator agreement or independent field validation (such as manual caliper measurements) is reported. In the revised manuscript we will add a dedicated subsection under Data Annotation that details the labeling workflow, the tools used, the number of annotators, any consistency checks performed, and the acknowledged limitations for fine structures such as trunks. This addition will allow readers to assess the reliability of the shared labels and the validity of the reported IoU comparisons. revision: yes

Circularity Check

0 steps flagged

No circularity; dataset release with standard baselines

full rationale

This is a data-release paper introducing the Viking Hill multi-sensor forest dataset with 3D cuboid annotations and reporting baseline semantic segmentation results using the off-the-shelf MinkowskiUNet model. No derivations, equations, predictions, or fitted parameters are presented that could reduce to the inputs by construction. The IoU numbers are direct outputs of applying a public segmentation network to the released point clouds; they are not claimed as novel theoretical results. No self-citation chains, uniqueness theorems, or ansatzes are invoked to support core claims. The work is therefore self-contained against external benchmarks and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a dataset collection paper with no mathematical derivations or new physical models. No free parameters, axioms beyond standard robotics sensor assumptions, or invented entities are introduced.

pith-pipeline@v0.9.1-grok · 5791 in / 1121 out tokens · 27841 ms · 2026-06-26T20:42:13.078667+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

47 extracted references · 8 canonical work pages

  1. [1]

    Forestry applications of UAVs in Europe: a review,

    C. Torresan, A. Berton, F. Carotenuto, S. F. Di Gennaro, B. Gioli, A. Matese, F. Miglietta, C. Vagnoli, A. Zaldei, and L. Wallace, “Forestry applications of UAVs in Europe: a review,”International Journal of Remote Sensing, vol. 38, no. 8-10, pp. 2427–2447, May 2017. [Online]. Available: https: //www.tandfonline.com/doi/full/10.1080/01431161.2016.1252477

    work page doi:10.1080/01431161.2016.1252477 2017

  2. [2]

    Advances in Forest Robotics: A State-of-the-Art Survey,

    L. F. P. Oliveira, A. P. Moreira, and M. F. Silva, “Advances in Forest Robotics: A State-of-the-Art Survey,”Robotics, vol. 10, no. 2, p. 53, Jun. 2021, publisher: Multidisciplinary Digital Publishing Institute. [Online]. Available: https://www.mdpi.com/2218-6581/10/2/53

    2021

  3. [3]

    Online Tree Reconstruction and Forest Inventory on a Mobile Robotic System,

    L. Freißmuth, M. Mattamala, N. Chebrolu, S. Schaefer, S. Leutenegger, and M. F. Fallon, “Online Tree Reconstruction and Forest Inventory on a Mobile Robotic System,” inIEEE/RSJ International Conference 27 on Intelligent Robots and Systems (IROS). Abu Dhabi, United Arab Emirates: IEEE, 2024, pp. 11 765–11 772. [Online]. Available: https://ieeexplore.ieee.o...

    arXiv 2024

  4. [4]

    Robotics in Forest Inventories: SPOT’s First Steps,

    G. Chirici, F. Giannetti, G. D’Amico, E. Vangi, S. Francini, C. Borghi, P. Corona, and D. Travaglini, “Robotics in Forest Inventories: SPOT’s First Steps,”Forests, vol. 14, no. 11, p. 2170, Oct. 2023. [Online]. Available: https://www.mdpi.com/1999-4907/14/11/2170

    2023

  5. [5]

    Exploring the feasibility of autonomous forestry operations: Results from the first experimental unmanned machine,

    P. La Hera, O. Mendoza-Trejo, and O. Lindroos, “Exploring the feasibility of autonomous forestry operations: Results from the first experimental unmanned machine,”Journal of Field Robotics, vol. 41, no. 4, pp. 942–965, 2024. [Online]. Available: https://onlinelibrary.wiley.com/doi/full/10.1002/rob.22300

    work page doi:10.1002/rob.22300 2024

  6. [6]

    Towards autonomous selective harvesting: A review of robot perception, robot design, motion planning and control,

    V. S. Rajendran, B. Debnath, S. Mghames, W. Mandil, S. Parsa, S. Parsons, and A. Ghalamzan-E, “Towards autonomous selective harvesting: A review of robot perception, robot design, motion planning and control,”Journal of Field Robotics, vol. 41, no. 7, pp. 2247–2279, 2024. [Online]. Available: https://onlinelibrary.wiley.com/ doi/full/10.1002/rob.22230

    work page doi:10.1002/rob.22230 2024

  7. [7]

    UAV-Assisted Self-Supervised Terrain Awareness for Off-Road Navigation,

    J.-M. Fortin, O. Gamache, W. Fecteau, E. Daum, W. Larriv´ ee-Hardy, F. Pomerleau, and P. Gigu` ere, “UAV-Assisted Self-Supervised Terrain Awareness for Off-Road Navigation,” in2025 IEEE International Conference on Robotics and Automation (ICRA), May 2025, pp. 1–7. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/ 11128050

    2025

  8. [8]

    Topological Mapping for Traversability-Aware Long- Range Navigation in Off-Road Terrain,

    J.-F. Tremblay, J. Alhosh, L. Petit, F. Lotfi, L. Landauro, and D. Meger, “Topological Mapping for Traversability-Aware Long- Range Navigation in Off-Road Terrain,” in2025 IEEE International Conference on Robotics and Automation (ICRA), May 2025, pp. 14 850–14 856. [Online]. Available: https://ieeexplore.ieee.org/ document/11128536

    arXiv 2025

  9. [9]

    Kilometer- scale autonomous navigation in subarctic forests: challenges and lessons learned,

    D. Baril, S.-P. Deschˆ enes, O. Gamache, M. Vaidis, D. LaRocque, J. Laconte, V. Kubelka, P. Gigu` ere, and F. Pomerleau, “Kilometer- scale autonomous navigation in subarctic forests: challenges and lessons learned,”Field Robotics, vol. 2, pp. 1628–1660, 2021. [Online]. Available: https://arxiv.org/pdf/2111.13981.pdf

    arXiv 2021

  10. [10]

    DigiForests: A Longitudinal LiDAR Dataset for Forestry Robotics,

    M. V. Malladi, N. Chebrolu, I. Scacchetti, L. Lobefaro, T. Guadagnino, B. Casseau, H. Oh, L. Freißmuth, M. Karppinen, J. Schweier, S. Leutenegger, J. Behley, C. Stachniss, and M. Fallon, “DigiForests: A Longitudinal LiDAR Dataset for Forestry Robotics,” inProc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA), 2025. 28

    2025

  11. [11]

    Obstacle Detection in Foliage with Ladar and Radar,

    L. Matthies, C. Bergh, A. Castano, J. Macedo, and R. Man- duchi, “Obstacle Detection in Foliage with Ladar and Radar,” in Robotics Research. The Eleventh International Symposium, P. Dario and R. Chatila, Eds. Berlin, Heidelberg: Springer, 2005, pp. 291–300

    2005

  12. [12]

    Radar Teach and Repeat: Architecture and Initial Field Testing,

    X. Qiao, A. Krawciw, S. Lilge, and T. D. Barfoot, “Radar Teach and Repeat: Architecture and Initial Field Testing,” in 2025 IEEE International Conference on Robotics and Automation (ICRA), May 2025, pp. 13 021–13 027. [Online]. Available: https: //ieeexplore.ieee.org/abstract/document/11128412

    arXiv 2025

  13. [13]

    RADIATE: A Radar Dataset for Automotive Perception in Bad Weather,

    M. Sheeny, E. De Pellegrin, S. Mukherjee, A. Ahrabian, S. Wang, and A. Wallace, “RADIATE: A Radar Dataset for Automotive Perception in Bad Weather,”2021 IEEE International Conference on Robotics and Automation (ICRA), pp. 1–7, May 2021, conference Name: 2021 IEEE International Conference on Robotics and Automation (ICRA) ISBN: 9781728190778 Place: Xi’an, ...

    arXiv 2021

  14. [14]

    Human Detection from 4D Radar Data in Low-Visibility Field Conditions,

    M. Skog, O. Kotlyar, V. Kubelka, and M. Magnusson, “Human Detection from 4D Radar Data in Low-Visibility Field Conditions,” Apr. 2024, arXiv:2404.05307 [cs]. [Online]. Available: http://arxiv. org/abs/2404.05307

    Pith/arXiv arXiv 2024

  15. [15]

    Xtonomy – autonomy everywhere,

    Xtonomy, “Xtonomy – autonomy everywhere,” 2026, accessed: 2026-06-10. [Online]. Available: https://www.xtonomy.ai/

    2026

  16. [16]

    A New Wave in Robotics: Survey on Recent MmWave Radar Applications in Robotics,

    K. Harlow, H. Jang, T. D. Barfoot, A. Kim, and C. Heckman, “A New Wave in Robotics: Survey on Recent MmWave Radar Applications in Robotics,”IEEE Transactions on Robotics, vol. 40, pp. 4544–4560, 2024. [Online]. Available: https://ieeexplore.ieee.org/ document/10683889

    arXiv 2024

  17. [17]

    Imaging radar for navigation and surveillance on an autonomous un- manned ground vehicle capable of detecting obstacles obscured by veg- etation,

    D. Gusland, B. Torvik, E. Finden, F. Gulbrandsen, and R. Smestad, “Imaging radar for navigation and surveillance on an autonomous un- manned ground vehicle capable of detecting obstacles obscured by veg- etation,” in2019 IEEE Radar Conference, RadarConf 2019, 2019

    2019

  18. [18]

    The GOOSE Dataset for Perception in Unstructured Environments,

    P. Mortimer, R. Hagmanns, M. Granero, T. Luettel, J. Petereit, and H.-J. Wuensche, “The GOOSE Dataset for Perception in Unstructured Environments,” in2024 IEEE International Conference on Robotics and Automation (ICRA), May 2024, pp. 14 838–14 844. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/10611298

    arXiv 2024

  19. [19]

    RELLIS- 3D Dataset: Data, Benchmarks and Analysis,

    P. Jiang, P. Osteen, M. Wigness, and S. Saripalli, “RELLIS- 3D Dataset: Data, Benchmarks and Analysis,” in2021 IEEE International Conference on Robotics and Automation (ICRA), 29 May 2021, pp. 1110–1116, iSSN: 2577-087X. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/9561251

    arXiv 2021

  20. [20]

    Tree detec- tion and diameter estimation based on deep learning,

    V. Grondin, J.-M. Fortin, F. Pomerleau, and P. Gigu` ere, “Tree detec- tion and diameter estimation based on deep learning,”Forestry: An International Journal of Forest Research, 10 2022

    2022

  21. [21]

    OORD: The Oxford Offroad Radar Dataset,

    M. Gadd, D. De Martini, O. Bartlett, P. Murcutt, M. Towlson, M. Widojo, V. Mu¸ sat, L. Robinson, E. Panagiotaki, G. Pramatarov, M. Alexander K¨ uhn, L. Marchegiani, P. Newman, and L. Kunze, “OORD: The Oxford Offroad Radar Dataset,”IEEE Transactions on Intelligent Transportation Systems, vol. 25, no. 11, pp. 18 779–18 790, Nov. 2024. [Online]. Available: h...

    arXiv 2024

  22. [22]

    The Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset,

    D. Barnes, M. Gadd, P. Murcutt, P. Newman, and I. Posner, “The Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset,” in2020 IEEE International Conference on Robotics and Automation (ICRA), May 2020, pp. 6433–6438, iSSN: 2577-087X. [Online]. Available: https://ieeexplore.ieee.org/document/9196884

    arXiv 2020

  23. [23]

    MulRan: Multimodal Range Dataset for Urban Place Recognition,

    G. Kim, Y. S. Park, Y. Cho, J. Jeong, and A. Kim, “MulRan: Multimodal Range Dataset for Urban Place Recognition,” in2020 IEEE International Conference on Robotics and Automation (ICRA), May 2020, pp. 6246–6253, iSSN: 2577-087X. [Online]. Available: https://ieeexplore.ieee.org/document/9197298

    arXiv 2020

  24. [24]

    Get It for Free: Radar Segmentation Without Expert Labels and Its Application in Odometry and Localization,

    S. Li, Z. Hong, Y. Chen, L. Hu, and J. Qin, “Get It for Free: Radar Segmentation Without Expert Labels and Its Application in Odometry and Localization,”IEEE Robotics and Automation Letters, vol. 10, no. 3, pp. 2678–2685, Mar. 2025. [Online]. Available: https://ieeexplore.ieee.org/document/10857300

    arXiv 2025

  25. [25]

    HeRCULES: Heterogeneous Radar Dataset in Complex Urban Environment for Multi-Session Radar SLAM,

    H. Kim, M. Jung, C. Noh, S. Jung, H. Song, W. Yang, H. Jang, and A. Kim, “HeRCULES: Heterogeneous Radar Dataset in Complex Urban Environment for Multi-Session Radar SLAM,” in2025 IEEE International Conference on Robotics and Automation (ICRA), May 2025, pp. 4649–4656. [Online]. Available: https: //ieeexplore.ieee.org/document/11128678

    arXiv 2025

  26. [26]

    4DRadarSLAM: A 4D Imaging Radar SLAM System for Large-scale Environments based on Pose Graph Optimization,

    J. Zhang, H. Zhuge, Z. Wu, G. Peng, M. Wen, Y. Liu, and D. Wang, “4DRadarSLAM: A 4D Imaging Radar SLAM System for Large-scale Environments based on Pose Graph Optimization,” in2023 IEEE International Conference on Robotics and Automation (ICRA), May 2023, pp. 8333–8340. [Online]. Available: https: //ieeexplore.ieee.org/abstract/document/10160670 30

    arXiv 2023

  27. [27]

    MSC-RAD4R: ROS-Based Automotive Dataset With 4D Radar,

    M. Choi, S. Yang, S. Han, Y. Lee, M. Lee, K. H. Choi, and K.-S. Kim, “MSC-RAD4R: ROS-Based Automotive Dataset With 4D Radar,”IEEE Robotics and Automation Letters, vol. 8, no. 11, pp. 7194–7201, Nov. 2023. [Online]. Available: https: //ieeexplore.ieee.org/document/10225273

    arXiv 2023

  28. [28]

    Dual Radar: A Multi-modal Dataset with Dual 4D Radar for Autononous Driving,

    X. Zhang, L. Wang, J. Chen, C. Fang, G. Yang, Y. Wang, L. Yang, Z. Song, L. Liu, X. Zhang, B. Xu, Z. Li, Q. Yang, J. Li, Z. Zhang, W. Wang, and S. S. Ge, “Dual Radar: A Multi-modal Dataset with Dual 4D Radar for Autononous Driving,”Sci Data, vol. 12, no. 1, p. 439, Mar. 2025, publisher: Nature Publishing Group. [Online]. Available: https://www.nature.com/...

    2025

  29. [29]

    K-Radar: 4D Radar Object Detection for Autonomous Driving in Various Weather Conditions,

    D.-H. Paek, S.-H. Kong, and K. T. Wijaya, “K-Radar: 4D Radar Object Detection for Autonomous Driving in Various Weather Conditions,” Jun. 2022. [Online]. Available: https: //openreview.net/forum?id=W bsDmzwaZ7

    2022

  30. [30]

    Multi-Class Road User Detection With 3+1D Radar in the View-of-Delft Dataset,

    A. Palffy, E. Pool, S. Baratam, J. F. P. Kooij, and D. M. Gavrila, “Multi-Class Road User Detection With 3+1D Radar in the View-of-Delft Dataset,”IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 4961–4968, Apr. 2022. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/9699098

    arXiv 2022

  31. [31]

    TJ4DRadSet: A 4D Radar Dataset for Autonomous Driving,

    L. Zheng, Z. Ma, X. Zhu, B. Tan, S. Li, K. Long, W. Sun, S. Chen, L. Zhang, M. Wan, L. Huang, and J. Bai, “TJ4DRadSet: A 4D Radar Dataset for Autonomous Driving,” in2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC), Oct. 2022, pp. 493–498. [Online]. Available: https: //ieeexplore.ieee.org/document/9922539

    arXiv 2022

  32. [32]

    nuScenes: A Multimodal Dataset for Autonomous Driving,

    H. Caesar, V. Bankiti, A. H. Lang, S. Vora, V. E. Liong, Q. Xu, A. Krishnan, Y. Pan, G. Baldan, and O. Beijbom, “nuScenes: A Multimodal Dataset for Autonomous Driving,” 2020, pp. 11 621–11 631. [Online]. Available: https://openaccess.thecvf.com/ content CVPR 2020/html/Caesar nuScenes A Multimodal Dataset for Autonomous Driving CVPR 2020 paper.html

    2020

  33. [33]

    ColoRadar: The direct 3D millimeter wave radar dataset,

    A. Kramer, K. Harlow, C. Williams, and C. Heckman, “ColoRadar: The direct 3D millimeter wave radar dataset,”The International Journal of Robotics Research, vol. 41, no. 4, pp. 351–360, Apr. 2022, publisher: SAGE Publications Ltd STM. [Online]. Available: https://doi.org/10.1177/02783649211068535

    work page doi:10.1177/02783649211068535 2022

  34. [34]

    CARRADA Dataset: Camera and Automotive Radar with Range- 31 Angle- Doppler Annotations,

    A. Ouaknine, A. Newson, J. Rebut, F. Tupin, and P. P´ erez, “CARRADA Dataset: Camera and Automotive Radar with Range- 31 Angle- Doppler Annotations,” in2020 25th International Conference on Pattern Recognition (ICPR), Jan. 2021, pp. 5068–5075, iSSN: 1051-4651. [Online]. Available: https://ieeexplore.ieee.org/document/ 9413181

    2021

  35. [35]

    Radar-Camera Fusion for Object Detection and Semantic Segmentation in Autonomous Driving: A Comprehensive Review,

    S. Yao, R. Guan, X. Huang, Z. Li, X. Sha, Y. Yue, E. G. Lim, H. Seo, K. L. Man, X. Zhu, and Y. Yue, “Radar-Camera Fusion for Object Detection and Semantic Segmentation in Autonomous Driving: A Comprehensive Review,”IEEE Transactions on Intelligent Vehicles, vol. 9, no. 1, pp. 2094–2128, Jan. 2024. [Online]. Available: https://ieeexplore.ieee.org/document/10225711

    arXiv 2094

  36. [36]

    Forest Aboveground Biomass Estimation and Inventory: Evaluating Remote Sensing-Based Approaches,

    M. N. Khan, Y. Tan, A. A. Gul, S. Abbas, and J. Wang, “Forest Aboveground Biomass Estimation and Inventory: Evaluating Remote Sensing-Based Approaches,”Forests, vol. 15, no. 6, p. 1055, Jun. 2024, publisher: Multidisciplinary Digital Publishing Institute. [Online]. Available: https://www.mdpi.com/1999-4907/15/6/1055

    2024

  37. [37]

    Overview and Benchmark on Multi-Modal Lidar Point Cloud Registration for Forest Applications,

    M. B. Campos, L. F. Castanheiro, D. Shah, Y. Wang, A. Kukko, and E. Puttonen, “Overview and Benchmark on Multi-Modal Lidar Point Cloud Registration for Forest Applications,”The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. XLVIII-1-2024, pp. 43–50, May 2024, conference Name: ISPRS TC I Mid-term Symposi...

    2024

  38. [38]

    Deep learning based 3D point cloud regression for estimating forest biomass,

    S. Oehmcke, L. Li, J. C. Revenga, T. Nord-Larsen, K. Trepekli, F. Gieseke, and C. Igel, “Deep learning based 3D point cloud regression for estimating forest biomass,” inProceedings of the 30th International Conference on Advances in Geographic Information Systems, ser. SIGSPATIAL ’22. New York, NY, USA: Association for Computing Machinery, Nov. 2022, pp. ...

    work page doi:10.1145/3557915.3561471 2022

  39. [39]

    Automatic three-dimensional mapping for tree diameter measurements in inventory operations,

    J.-F. Tremblay, M. B´ eland, R. Gagnon, F. Pomerleau, and P. Gigu` ere, “Automatic three-dimensional mapping for tree diameter measurements in inventory operations,”Journal of Field Robotics, vol. 37, no. 8, pp. 1328–1346, 2020, eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/rob.21980. [Online]. Available: https://onlinelibrary.wiley.com/doi/abs/...

    work page doi:10.1002/rob.21980 2020

  40. [40]

    FOR-instance: a UAV laser scanning benchmark dataset for semantic and instance segmentation of 32 individual trees,

    S. Puliti, G. Pearse, P. Surov´ y, L. Wallace, M. Hollaus, M. Wielgosz, and R. Astrup, “FOR-instance: a UAV laser scanning benchmark dataset for semantic and instance segmentation of 32 individual trees,” Sep. 2023, arXiv:2309.01279 [cs]. [Online]. Available: http://arxiv.org/abs/2309.01279

    arXiv 2023

  41. [41]

    TreeScope: An Agricultural Robotics Dataset for LiDAR-Based Mapping of Trees in Forests and Orchards,

    D. Cheng, F. Cladera, A. Prabhu, X. Liu, A. Zhu, P. C. Green, R. Ehsani, P. Chaudhari, and V. Kumar, “TreeScope: An Agricultural Robotics Dataset for LiDAR-Based Mapping of Trees in Forests and Orchards,” in2024 IEEE International Conference on Robotics and Automation (ICRA), May 2024, pp. 14 860–14 866. [Online]. Available: https://ieeexplore.ieee.org/ab...

    arXiv 2024

  42. [42]

    WildScenes: A benchmark for 2D and 3D semantic segmentation in large-scale natural environments,

    K. Vidanapathirana, J. Knights, S. Hausler, M. Cox, M. Ramezani, J. Jooste, E. Griffiths, S. Mohamed, S. Sridharan, C. Fookes, and P. Moghadam, “WildScenes: A benchmark for 2D and 3D semantic segmentation in large-scale natural environments,”The International Journal of Robotics Research, vol. 44, no. 4, pp. 532–549, Apr. 2025, publisher: SAGE Publication...

    work page doi:10.1177/02783649241278369 2025

  43. [43]

    4d spatio-temporal convnets: Minkowski convolutional neural networks,

    C. Choy, J. Gwak, and S. Savarese, “4d spatio-temporal convnets: Minkowski convolutional neural networks,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2019, pp. 3075–3084

    2019

  44. [44]

    Microsoft COCO: Common Objects in Context,

    T.-Y. Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Doll´ ar, and C. L. Zitnick, “Microsoft COCO: Common Objects in Context,” inComputer Vision – ECCV 2014, D. Fleet, T. Pajdla, B. Schiele, and T. Tuytelaars, Eds. Cham: Springer International Publishing, 2014, pp. 740–755

    2014

  45. [45]

    Egoblur: Responsible innovation in aria,

    N. Raina, G. Somasundaram, K. Zheng, S. Miglani, S. Saarinen, J. Meissner, M. Schwesinger, L. Pesqueira, I. Prasad, E. Miller, P. Gupta, M. Yan, R. Newcombe, C. Ren, and O. M. Parkhi, “Egoblur: Responsible innovation in aria,” 2023

    2023

  46. [46]

    Comparing ICP Variants on Real-World Data Sets,

    F. Pomerleau, F. Colas, R. Siegwart, and S. Magnenat, “Comparing ICP Variants on Real-World Data Sets,”Autonomous Robots, vol. 34, no. 3, pp. 133–148, Feb. 2013

    2013

  47. [47]

    A portable three- dimensional LIDAR-based system for long-term and wide-area people behavior measurement,

    K. Koide, J. Miura, and E. Menegatti, “A portable three- dimensional LIDAR-based system for long-term and wide-area people behavior measurement,”International Journal of Advanced Robotic Systems, vol. 16, no. 2, p. 1729881419841532, Mar. 2019, publisher: SAGE Publications. [Online]. Available: https: //doi.org/10.1177/1729881419841532 33

    work page doi:10.1177/1729881419841532 2019