pith. sign in

arxiv: 2606.32023 · v1 · pith:GBIVOEUWnew · submitted 2026-06-30 · 💻 cs.CV · cs.AI

FLORA: A deep learning approach to predict forest attributes from heterogeneous LiDAR data

Emilie Vautier , Cl\'ement Mallet , C\'edric Vega This is my paper

Pith reviewed 2026-07-01 05:29 UTC · model grok-4.3

classification 💻 cs.CV cs.AI
keywords deep learningLiDAR point cloudsforest attributesoctree regressionnational forest inventoryheterogeneous datacross-season robustnessauxiliary variables
0
0 comments X

The pith

A single model trained on mixed leaf-on and leaf-off LiDAR outperforms season-specific models for forest attributes.

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

The paper presents FLORA, a deep learning method that processes heterogeneous LiDAR point clouds to predict six forest attributes including dominant height and volumes. It demonstrates that training one model on both seasonal acquisitions yields better cross-season performance than separate models calibrated to leaf-on or leaf-off data alone. This addresses the challenge of scaling predictions across national LiDAR programs where sensors, flight parameters, and seasons vary. Results on 32,052 plots show rRMSE of 12.3 percent for dominant height and 39 percent for total volume, establishing a baseline for large-area estimation.

Core claim

FLORA combines an octree-based backbone with ecological and spatiotemporal auxiliary variables through a late-fusion gating mechanism. Models trained and evaluated on 32,052 National Forest Inventory plots across mainland France show that a single model trained on both leaf-on and leaf-off acquisitions outperforms season-specific models and improves cross-season robustness. Auxiliary variables provide modest overall gains but contribute more strongly to species-specific volume prediction. FLORA achieves an rRMSE of about 12.3 percent (R2 = 0.88) for dominant height and 39 percent (R2 = 0.74) for total volume.

What carries the argument

Octree-based backbone with late-fusion gating mechanism that merges LiDAR point clouds and auxiliary variables for attribute regression.

If this is right

  • Wall-to-wall mapping of forest attributes becomes feasible across entire national territories despite acquisition inconsistencies.
  • Species-specific volume estimates gain from auxiliary data even when overall gains remain modest.
  • Cross-season training reduces the need to maintain separate models for different LiDAR campaigns.
  • The reported error levels serve as a performance reference for other European national LiDAR programs.

Where Pith is reading between the lines

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

  • The approach could extend to countries with comparable LiDAR programs if similar plot networks and auxiliary layers exist.
  • Octree processing may offer a general route for handling variable-density point clouds in other remote-sensing tasks beyond forests.
  • Further gains might appear if auxiliary variables are expanded to include more detailed soil or climate layers.

Load-bearing premise

The 32,052 NFI plots and chosen auxiliary variables are sufficient to capture and correct for the full range of sensor, flight, season, and scan-angle variability in the French LiDAR program.

What would settle it

Testing the single mixed-season model against season-specific models on a new set of plots from unseen sensors or acquisition conditions and finding equal or lower accuracy for the mixed model.

Figures

Figures reproduced from arXiv: 2606.32023 by C\'edric Vega, Cl\'ement Mallet, Emilie Vautier.

Figure 1
Figure 1. Figure 1: Variability of our plot-level LiDAR dataset in terms of phenology, sensor, scan [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Spatial distribution over France of the NFI plots used in this study, colored with [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Step-wise description of FLORA. ch stands for channels, and indicates the size of the feature space at each stage. All provided numbers correspond to the real values for the illustrated plot. 14 [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Comparison of predictive performance between the full model (solid black bars) [PITH_FULL_IMAGE:figures/full_fig_p024_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Effect of spatial overlap between LiDAR acquisitions on model performance. Met [PITH_FULL_IMAGE:figures/full_fig_p027_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Barplots showing changes in rRMSE and R2 between the LiDAR-only baseline (No Aux) and FLORA (Aux). Impact on predictive accuracy ( [PITH_FULL_IMAGE:figures/full_fig_p028_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Distribution of learned gating coefficients for each auxiliary data configuration [PITH_FULL_IMAGE:figures/full_fig_p029_7.png] view at source ↗
read the original abstract

Forest attributes are essential for national-scale resource monitoring. Airborne LiDAR metrics are among the auxiliary variables most strongly correlated with forest attributes used in National Forest Inventory (NFI) estimates. However, producing wall-to-wall predictions remains challenging when LiDAR data are acquired under heterogeneous conditions. As national LiDAR programs expand across Europe, variability in sensors, flight parameters, seasons, and scan angles limits the robustness of existing models, which are often calibrated for local conditions. We present FLORA (Forest LiDAR Octree Regression with Auxiliary Data), a deep learning framework that predicts six forest attributes: dominant height, total volume, deciduous volume, coniferous volume, basal area, and stem density from heterogeneous LiDAR point clouds. FLORA combines an octree-based backbone with ecological and spatiotemporal auxiliary variables through a late-fusion gating mechanism. Models are trained and evaluated on 32,052 National Forest Inventory plots across mainland France using data from the French LiDAR HD program. A single model trained on both leaf-on and leaf-off acquisitions outperforms season-specific models and improves cross-season robustness. Auxiliary variables provide modest overall gains but contribute more strongly to species-specific volume prediction. FLORA achieves an rRMSE of about 12.3% (R2 = 0.88) for dominant height and 39% (R2 = 0.74) for total volume, providing a robust baseline for large-scale forest attribute estimation from heterogeneous national LiDAR programs.

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

2 major / 2 minor

Summary. The manuscript presents FLORA, a deep learning framework that uses an octree-based backbone combined with ecological and spatiotemporal auxiliary variables through late-fusion gating to predict six forest attributes (dominant height, total volume, deciduous/coniferous volume, basal area, stem density) from heterogeneous airborne LiDAR point clouds. Models are trained and evaluated on 32,052 National Forest Inventory plots across France using data from the French LiDAR HD program. The central claim is that a single model trained on mixed leaf-on and leaf-off acquisitions outperforms season-specific models, improves cross-season robustness, and achieves rRMSE of ~12.3% (R²=0.88) for dominant height and 39% (R²=0.74) for total volume, serving as a baseline for large-scale estimation under variable national LiDAR conditions.

Significance. If the robustness and performance claims hold after addressing validation details, the work addresses a practical barrier to wall-to-wall forest attribute mapping as European national LiDAR programs scale up. The large plot sample and explicit handling of season heterogeneity via a joint model represent a concrete advance over locally calibrated approaches; the auxiliary-variable contribution to species-specific volumes is a useful secondary finding.

major comments (2)
  1. [Abstract] The claim that the mixed-season model improves cross-season robustness (Abstract) is load-bearing for the paper's contribution, yet the manuscript provides no stratification or coverage analysis showing that all major sensor/flight/season/scan-angle combinations appear with adequate sample size in both training and test partitions. Without this, the reported rRMSE gains could reflect the observed distribution rather than genuine generalization across the full LiDAR HD heterogeneity.
  2. The reported performance metrics (rRMSE 12.3% for height, 39% for volume) are presented without accompanying details on the cross-validation strategy, train/test split independence, error bars, or data exclusion rules. These omissions make it impossible to determine whether the numbers reflect truly held-out evaluation or optimistic in-sample performance.
minor comments (2)
  1. [Abstract] The abstract states that auxiliary variables 'provide modest overall gains' but does not quantify the incremental improvement (e.g., via ablation tables) or specify which variables drive the species-specific volume benefit.
  2. Notation for the six target attributes and the exact definition of rRMSE should be stated explicitly in the methods or results section for reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback highlighting the need for clearer validation details to support the generalization claims. We address each major comment below and will revise the manuscript to incorporate the requested clarifications.

read point-by-point responses

  1. Referee: [Abstract] The claim that the mixed-season model improves cross-season robustness (Abstract) is load-bearing for the paper's contribution, yet the manuscript provides no stratification or coverage analysis showing that all major sensor/flight/season/scan-angle combinations appear with adequate sample size in both training and test partitions. Without this, the reported rRMSE gains could reflect the observed distribution rather than genuine generalization across the full LiDAR HD heterogeneity.

    Authors: We agree that explicit coverage analysis is necessary to substantiate the cross-season robustness claim. In the revised manuscript we will add a new subsection (and accompanying table) in the Methods that reports the counts and proportions of leaf-on vs. leaf-off acquisitions, sensor models, and binned scan-angle ranges within both the training and test partitions. This will demonstrate that the major heterogeneity factors are represented in both sets with sufficient sample sizes. revision: yes

  2. Referee: [—] The reported performance metrics (rRMSE 12.3% for height, 39% for volume) are presented without accompanying details on the cross-validation strategy, train/test split independence, error bars, or data exclusion rules. These omissions make it impossible to determine whether the numbers reflect truly held-out evaluation or optimistic in-sample performance.

    Authors: We acknowledge that these methodological details were insufficiently described. The revised Methods section will explicitly state that a 5-fold spatial cross-validation was used, with folds constructed to ensure geographic separation of plots (minimum 5 km buffer between train and test regions) to maintain independence. We will also report standard deviations across the five folds as error bars on all metrics and list the precise exclusion criteria applied to the 32 052 plots. revision: yes

Circularity Check

0 steps flagged

No circularity detected; empirical results measured on held-out plots

full rationale

The paper describes an empirical deep learning pipeline (octree backbone + late-fusion auxiliaries) trained and evaluated on 32,052 NFI plots with performance reported against independent field measurements. No equations, derivations, or self-citations appear in the provided text that reduce any claimed result to its inputs by construction. The central claims rest on standard supervised learning against external benchmarks rather than self-definitional steps, fitted-input predictions, or load-bearing self-citations.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the representativeness of the French NFI plots for heterogeneous conditions and on the capacity of the learned gating mechanism to compensate for sensor and acquisition variability; these are not independently verified outside the training data.

free parameters (1)
  • neural network parameters
    All weights in the octree backbone and gating network are fitted to the 32,052 plots.
axioms (1)
  • domain assumption The selected auxiliary variables and NFI plots adequately represent the variability in sensors, flight parameters, seasons, and scan angles across the French LiDAR HD program.
    Invoked when claiming cross-season robustness and national-scale applicability.

pith-pipeline@v0.9.1-grok · 5801 in / 1313 out tokens · 26655 ms · 2026-07-01T05:29:34.409471+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

108 extracted references · 97 canonical work pages · 3 internal anchors

  1. [2]

    and Ye, Jong Chul , month = jan, year =

    Chung, Hyungjin and Lee, Dohun and Wu, Zihui and Kim, Byung-Hoon and Bouman, Katherine L. and Ye, Jong Chul , month = jan, year =. doi:10.48550/arXiv.2501.04284 , abstract =

    work page doi:10.48550/arxiv.2501.04284

  2. [3]

    Akiba, Takuya and Shotaro, Sano and Toshihiko, Yanase and Takeru, Ohta and Masanori, Koyama , year =. Optuna:

  3. [4]

    2022 , note =

    Remote Sensing , author =. 2022 , note =. doi:10.3390/rs14153825 , abstract =

    work page doi:10.3390/rs14153825 2022

  4. [5]

    Speeding up common hyperparameter optimization methods by a two-phase-search , copyright =

    Wendt, Alexander and Wuschnig, Marco and Lechner, Martin , month = oct, year =. Speeding up common hyperparameter optimization methods by a two-phase-search , copyright =. doi:10.1109/IECON43393.2020.9254801 , urldate =

    work page doi:10.1109/iecon43393.2020.9254801 2020

  5. [6]

    2023 , pages =

    ACM Transactions on Graphics , author =. 2023 , pages =. doi:10.1145/3592131 , abstract =

    work page doi:10.1145/3592131 2023

  6. [13]

    Proceedings of the

    Gaydon, Charles and Roche, Floryne , year =. Proceedings of the

  7. [14]

    Gaydon, Charles , year =

  8. [15]

    Open-canopy: towards very high resolution forest monitoring , booktitle =

    Fogel, Fajwel and Perron, Yohann and Besic, Nikola and Saint-André, Laurent and Pellissier-Tanon, Agnès and Schwartz, Martin and Boudras, Thoma and Fayad, Ibrahim and d'Aspremont, Alexandre and Landrieu, Loic and Ciais, Philippe , year =. Open-canopy: towards very high resolution forest monitoring , booktitle =

  9. [16]

    Biotropica , author =

    The definition of treefall gap and its effect on measures of forest dynamics , volume =. Biotropica , author =. 1982 , note =. doi:10.2307/2387750 , number =

    work page doi:10.2307/2387750 1982

  10. [18]

    Algorithms for hyper-parameter optimization , volume =

    Bergstra, James and Bardenet, Rémi and Bengio, Yoshua and Kégl, Balázs , editor =. Algorithms for hyper-parameter optimization , volume =. Advances in

  11. [19]

    Northern Journal of Applied Forestry , author =

    Estimating aboveground biomass and average annual wood biomass increment with airborne leaf-on and leaf-off. Northern Journal of Applied Forestry , author =. 2013 , pages =. doi:10.5849/njaf.12-015 , abstract =

    work page doi:10.5849/njaf.12-015 2013

  12. [20]

    Artificial Intelligence for the Earth Systems , author =

    Two-step hyperparameter optimization method: accelerating hyperparameter search by using a fraction of a training dataset , volume =. Artificial Intelligence for the Earth Systems , author =. 2024 , note =. doi:10.1175/AIES-D-23-0013.1 , abstract =

    work page doi:10.1175/aies-d-23-0013.1 2024

  13. [22]

    What makes training multi-modal classification networks hard? , url =

    Wang, Weiyao and Tran, Du and Feiszli, Matt , month = apr, year =. What makes training multi-modal classification networks hard? , url =. doi:10.48550/arXiv.1905.12681 , abstract =

    work page doi:10.48550/arxiv.1905.12681 1905

  14. [24]

    Forests , author =

    Review of remote sensing-based methods for forest aboveground biomass estimation: progress, challenges, and prospects , volume =. Forests , author =. 2023 , note =. doi:10.3390/f14061086 , abstract =

    work page doi:10.3390/f14061086 2023

  15. [27]

    2009 , note =

    International Journal of Forestry Research , author =. 2009 , note =. doi:10.1155/2009/864108 , abstract =

    work page doi:10.1155/2009/864108 2009

  16. [32]

    Li, Ruihui and Li, Xianzhi and Heng, Pheng-Ann and Fu, Chi-Wing , month = jun, year =. 2020. doi:10.1109/cvpr42600.2020.00641 , abstract =

    work page doi:10.1109/cvpr42600.2020.00641 2020

  17. [33]

    Current Forestry Reports , author =

    Artificial intelligence and terrestrial point clouds for forest monitoring , volume =. Current Forestry Reports , author =. 2024 , keywords =. doi:10.1007/s40725-024-00234-4 , abstract =

    work page doi:10.1007/s40725-024-00234-4 2024

  18. [37]

    IEEE Access , author =

    Cross-subject. IEEE Access , author =. 2019 , keywords =. doi:10.1109/ACCESS.2019.2939288 , abstract =

    work page doi:10.1109/access.2019.2939288 2019

  19. [40]

    Nature Machine Intelligence , author =

    Shortcut learning in deep neural networks , volume =. Nature Machine Intelligence , author =. 2020 , note =. doi:10.1038/s42256-020-00257-z , abstract =

    work page doi:10.1038/s42256-020-00257-z 2020

  20. [42]

    The unreasonable effectiveness of early discarding after one epoch in neural network hyperparameter optimization , url =

    Egele, Romain and Mohr, Felix and Viering, Tom and Balaprakash, Prasanna , month = apr, year =. The unreasonable effectiveness of early discarding after one epoch in neural network hyperparameter optimization , url =. doi:10.48550/arXiv.2404.04111 , abstract =

    work page doi:10.48550/arxiv.2404.04111

  21. [43]

    Complexity , author =

    Effect of the sampling of a dataset in the hyperparameter optimization phase over the efficiency of a machine learning algorithm , volume =. Complexity , author =. 2019 , note =. doi:10.1155/2019/6278908 , abstract =

    work page doi:10.1155/2019/6278908 2019

  22. [44]

    Applied Sciences , author =

    Semi-supervised domain adaptation for individual identification from electrocardiogram signals , volume =. Applied Sciences , author =. 2023 , note =. doi:10.3390/app132413259 , abstract =

    work page doi:10.3390/app132413259 2023

  23. [47]

    Remote Sensing , author =

    The use of three-dimensional convolutional neural networks to interpret. Remote Sensing , author =. 2018 , note =. doi:10.3390/rs10040649 , abstract =

    work page doi:10.3390/rs10040649 2018

  24. [56]

    Remote Sensing of Environment , author =

    Retrieving yearly forest growth from satellite data:. Remote Sensing of Environment , author =. 2025 , pages =. doi:10.1016/j.rse.2025.114959 , language =

    work page doi:10.1016/j.rse.2025.114959 2025

  25. [59]

    ISPRS Journal of Photogrammetry and Remote Sensing , author =

    Harnessing conditional generative adversarial networks for. ISPRS Journal of Photogrammetry and Remote Sensing , author =. 2026 , pages =. doi:10.1016/j.isprsjprs.2026.01.043 , language =

    work page doi:10.1016/j.isprsjprs.2026.01.043 2026

  26. [71]

    Science of Remote Sensing , author =

    Bridging spatio-temporal gaps in. Science of Remote Sensing , author =. 2025 , pages =. doi:10.1016/j.srs.2025.100318 , abstract =

    work page doi:10.1016/j.srs.2025.100318 2025

  27. [73]

    calameo.com , author =

    Mémento inventaire forestier. calameo.com , author =

  28. [75]

    2026 , pages =

    ISPRS Journal of Photogrammetry and Remote Sensing , author =. 2026 , pages =. doi:10.1016/j.isprsjprs.2026.02.002 , language =

    work page doi:10.1016/j.isprsjprs.2026.02.002 2026

  29. [76]

    Scientific Reports , author =

    An aerial point cloud classification using point transformer via multi-feature fusion , volume =. Scientific Reports , author =. 2025 , pages =. doi:10.1038/s41598-025-02719-z , language =

    work page doi:10.1038/s41598-025-02719-z 2025

  30. [80]

    Canadian Journal of Forest Research , author =

    Harmonizing multi-temporal airborne laser scanning point clouds to derive periodic annual height increments in temperate mixedwood forests , volume =. Canadian Journal of Forest Research , author =. 2022 , note =. doi:10.1139/cjfr-2022-0055 , abstract =

    work page doi:10.1139/cjfr-2022-0055 2022

  31. [83]

    Silva Fennica , author =

    The suitability of leaf-off airborne laser scanning data in an area-based forest inventory of coniferous and deciduous trees , volume =. Silva Fennica , author =

  32. [88]

    Geophysical Research Letters , author =

    Biomass estimation in a tropical wet forest using. Geophysical Research Letters , author =. 2010 , note =. doi:10.1029/2010GL045608 , abstract =

    work page doi:10.1029/2010gl045608 2010

  33. [89]

    Optics Express , author =

    Estimation of coniferous forest aboveground biomass with aggregated airborne small-footprint. Optics Express , author =. 2017 , pages =. doi:10.1364/OE.25.00A851 , abstract =

    work page doi:10.1364/oe.25.00a851 2017

  34. [90]

    Remote Sensing of Environment , author =

    Automated forest inventory:. Remote Sensing of Environment , author =. 2024 , pages =. doi:10.1016/j.rse.2024.114078 , abstract =

    work page doi:10.1016/j.rse.2024.114078 2024

  35. [93]

    Remote Sensing of Environment , author =

    Generalizing predictive models of forest inventory attributes using an area-based approach with airborne. Remote Sensing of Environment , author =. 2015 , pages =. doi:10.1016/j.rse.2014.10.004 , language =

    work page doi:10.1016/j.rse.2014.10.004 2015

  36. [95]

    Remote Sensing of Environment , author =

    Hybrid three-phase estimators for large-area forest inventory using ground plots, airborne lidar, and space lidar , volume =. Remote Sensing of Environment , author =. 2017 , pages =. doi:10.1016/j.rse.2017.04.004 , language =

    work page doi:10.1016/j.rse.2017.04.004 2017

  37. [96]

    Remote Sensing of Environment , author =

    Measurement of fine-spatial-resolution. Remote Sensing of Environment , author =. 2017 , pages =. doi:10.1016/j.rse.2016.10.041 , abstract =

    work page doi:10.1016/j.rse.2016.10.041 2017

  38. [97]

    ISPRS Journal of Photogrammetry and Remote Sensing , author =

    Full-waveform topographic lidar:. ISPRS Journal of Photogrammetry and Remote Sensing , author =. 2009 , pages =. doi:10.1016/j.isprsjprs.2008.09.007 , abstract =

    work page doi:10.1016/j.isprsjprs.2008.09.007 2009

  39. [98]

    Agricultural and Forest Meteorology , author =

    Unlocking the potential of. Agricultural and Forest Meteorology , author =. 2025 , pages =. doi:10.1016/j.agrformet.2024.110341 , abstract =

    work page doi:10.1016/j.agrformet.2024.110341 2025

  40. [99]

    Remote Sensing of Environment , author =

    Statistical rigor in. Remote Sensing of Environment , author =. 2016 , pages =. doi:10.1016/j.rse.2015.11.012 , language =

    work page doi:10.1016/j.rse.2015.11.012 2016

  41. [100]

    Remote Sensing of Environment , author =

    Statistically rigorous, model-based inferences from maps , volume =. Remote Sensing of Environment , author =. 2022 , pages =. doi:10.1016/j.rse.2022.113028 , abstract =

    work page doi:10.1016/j.rse.2022.113028 2022

  42. [101]

    Annals of Forest Science , author =

    Band configurations and seasonality influence the predictions of common boreal tree species using. Annals of Forest Science , author =. 2024 , note =. doi:10.1186/s13595-024-01251-w , abstract =

    work page doi:10.1186/s13595-024-01251-w 2024

  43. [103]

    , author Shotaro, S

    author Akiba, T. , author Shotaro, S. , author Toshihiko, Y. , author Takeru, O. , author Masanori, K. , year 2019 . title Optuna: A next-generation hyperparameter optimization framework . https://dl.acm.org/doi/epdf/10.1145/3292500.3330701, :10.1145/3292500.3330701

    work page doi:10.1145/3292500.3330701 2019

  44. [104]

    , author Hayes, D.J

    author Ayrey, E. , author Hayes, D.J. , author Kilbride, J.B. , author Fraver, S. , author John A. Kershaw, J. , author Cook, B.D. , author Weiskittel, A.R. , year 2021 . title Synthesizing disparate LiDAR and satellite datasets through deep learning to generate wall-to-wall regional inventories for the complex, mixed-species forests of the eastern United...

    work page doi:10.3390/rs13245113 2021

  45. [105]

    , author Bardenet, R

    author Bergstra, J. , author Bardenet, R. , author Bengio, Y. , author Kégl, B. , year 2011 . title Algorithms for hyper-parameter optimization , in: editor Shawe-Taylor, J. , editor Zemel, R. , editor Bartlett, P. , editor Pereira, F. , editor Weinberger, K.Q. (Eds.), booktitle Advances in Neural Information Processing Systems , publisher Curran Associates, Inc

    2011

  46. [106]

    , author Ericsson, M

    author Björnberg, D. , author Ericsson, M. , author Lindeberg, J. , author Löwe, W. , author Nordqvist, J. , author Wallerman, J. , author Fransson, J.E. , year 2026 . title Improving national forest attribute maps of Sweden with machine learning . journal Science of Remote Sensing volume 13 , pages 100395 . https://linkinghub.elsevier.com/retrieve/pii/S2...

    work page doi:10.1016/j.srs.2026.100395 2026

  47. [107]

    , author Hervé, J.C

    author Bontemps, J.D. , author Hervé, J.C. , author Denardou, A. , year 2019 . title Partition idéalisée et régionalisée de la composition en espèces ligneuses des forêts françaises . journal Écoscience volume 26 , pages 291--308 . https://doi.org/10.1080/11956860.2019.1588511, :10.1080/11956860.2019.1588511. note publisher: Taylor & Francis \_eprint: htt...

    work page doi:10.1080/11956860.2019.1588511 2019

  48. [108]

    , author Abegg, M

    author Bornand, A. , author Abegg, M. , author Morsdorf, F. , author Rehush, N. , year 2024 . title Completing 3D point clouds of individual trees using deep learning . journal Methods in Ecology and Evolution volume 15 , pages 2010--2023 . https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14412, :10.1111/2041-210X.14412

    work page doi:10.1111/2041-210x.14412 2024

  49. [109]

    , author Nazeer, M

    author Borsah, A.A. , author Nazeer, M. , author Wong, M.S. , year 2023 . title LIDAR -based forest biomass remote sensing: a review of metrics, methods, and assessment criteria for the selection of allometric equations . journal Forests volume 14 . :10.3390/f14102095

    work page doi:10.3390/f14102095 2023

  50. [110]

    , author Antón-Fernández, C

    author Breidenbach, J. , author Antón-Fernández, C. , author Petersson, H. , author McRoberts, R.E. , author Astrup, R. , year 2014 . title Quantifying the model-relatedvariability of biomass stock and change estimates in the Norwegian national forest inventory . journal Forest Science volume 60 , pages 25--33 . https://doi.org/10.5849/forsci.12-137, :10....

    work page doi:10.5849/forsci.12-137 2014

  51. [111]

    , author Tompalski, P

    author Coops, N. , author Tompalski, P. , author Goodbody, T. , author Queinnec, M. , author Luther, J. , author Bolton, D. , author White, J. , author Wulder, M. , author Lier, O. , author Hermosilla, T. , year 2021 . title Modelling lidar-derived estimates of forest attributes over space and time: A review of approaches and future trends . journal Remot...

    work page doi:10.1016/j.rse.2021.112477 2021

  52. [112]

    , author Donoghue, D.N.M

    author Davison, S. , author Donoghue, D.N.M. , author Galiatsatos, N. , year 2020 . title The effect of leaf-on and leaf-off forest canopy conditions on LiDAR derived estimations of forest structural diversity . journal International Journal of Applied Earth Observation and Geoinformation volume 92 , pages 102160 . https://www.sciencedirect.com/science/ar...

    work page doi:10.1016/j.jag.2020.102160 2020

  53. [113]

    , author Knoke, T

    author Fan, W. , author Knoke, T. , author Troles, J. , author Tian, J. , year 2026 . title Automatic tree-level based forest inventories retrieval via ultra-high resolution UAV images and deep learning . journal ISPRS Journal of Photogrammetry and Remote Sensing volume 234 , pages 261--274 . https://linkinghub.elsevier.com/retrieve/pii/S0924271626000882,...

    work page doi:10.1016/j.isprsjprs.2026.02.029 2026

  54. [114]

    , author Silva, C.A

    author Fareed, N. , author Silva, C.A. , author Numata, I. , author Flores, J.P. , year 2026 . title Interdisciplinary applications of LiDAR in forest studies: advances in sensors, methods, and cross-domain metrics . journal Remote Sensing volume 18 . :10.3390/rs18020219

    work page doi:10.3390/rs18020219 2026

  55. [115]

    , year 2022

    author Gaydon, C. , year 2022 . title Myria3D : deep learning for the semantic segmentation of aerial Lidar point clouds . https://github.com/IGNF/myria3d. note original-date: 2022-01-10T09:50:28Z

    2022

  56. [116]

    , author Roche, F

    author Gaydon, C. , author Roche, F. , year 2025 . title PureForest : A large-scale aerial Lidar and aerial imagery dataset for tree species classification in monospecific forests , in: booktitle Proceedings of the Winter Conference on Applications of Computer Vision ( WACV ) , pp. pages 5895--5904

    2025

  57. [117]

    , author Landrieu, L

    author Geist, L. , author Landrieu, L. , author Robert, D. , year 2025 . title EZ - SP : fast and lightweight superpoint-based 3D segmentation . http://arxiv.org/abs/2512.00385, :10.48550/arXiv.2512.00385. note arXiv:2512.00385 [cs]

    work page doi:10.48550/arxiv.2512.00385 2025

  58. [118]

    , author Thomas, V.A

    author Gopalakrishnan, R. , author Thomas, V.A. , author Coulston, J.W. , author Wynne, R.H. , year 2015 . title Prediction of canopy heights over a large region using heterogeneous Lidar datasets: efficacy and challenges . journal Remote Sensing :10.3390/rs70911036

    work page doi:10.3390/rs70911036 2015

  59. [119]

    , author Manka, F

    author Guindon, L. , author Manka, F. , author Correia, D.L. , author Villemaire, P. , author Smiley, B. , author Bernier, P. , author Gauthier, S. , author Beaudoin, A. , author Boucher, J. , author Boulanger, Y. , year 2024 . title A new approach for spatializing the Canadian National Forest Inventory ( SCANFI ) using Landsat dense time series . journal...

    work page doi:10.1139/cjfr-2023-0118 2024

  60. [120]

    , author Rahlf, J

    author Hauglin, M. , author Rahlf, J. , author Schumacher, J. , author Astrup, R. , author Breidenbach, J. , year 2021 . title Large scale mapping of forest attributes using heterogeneous sets of airborne laser scanning and National Forest Inventory data . journal Forest Ecosystems volume 8 , pages 65 . https://www.sciencedirect.com/science/article/pii/S2...

    work page doi:10.1186/s40663-021-00338-4 2021

  61. [121]

    , author Francini, S

    author Hawryło, P. , author Francini, S. , author Chirici, G. , author Giannetti, F. , author Parkitna, K. , author Krok, G. , author Mitelsztedt, K. , author Lisańczuk, M. , author Stereńczak, K. , author Ciesielski, M. , author Wężyk, P. , author Socha, J. , year 2020 . title The use of remotely sensed data and Polish NFI plots for prediction of growing...

    work page doi:10.3390/rs12203331 2020

  62. [122]

    title Mémento inventaire forestier IGN - 2025

    author IGN , year 2025 . title Mémento inventaire forestier IGN - 2025 . https://www.calameo.com/read/0011885827c418636cbd4

    work page arXiv 2025

  63. [123]

    , author Renaud, J.P

    author Irulappa-Pillai-Vijayakumar, D.B. , author Renaud, J.P. , author Morneau, F. , author McRoberts, R.E. , author Vega, C. , year 2019 . title Increasing precision for french forest inventory estimates using the k- NN technique with optical and photogrammetric data and model-assisted estimators . journal Remote Sensing volume 11 . https://www.mdpi.com...

    work page doi:10.3390/rs11080991 2019

  64. [124]

    , author Guo, Q

    author Jakubowski, M.K. , author Guo, Q. , author Kelly, M. , year 2013 . title Tradeoffs between lidar pulse density and forest measurement accuracy . journal Remote Sensing of Environment volume 130 , pages 245--253 . https://linkinghub.elsevier.com/retrieve/pii/S0034425712004567, :10.1016/j.rse.2012.11.024

    work page doi:10.1016/j.rse.2012.11.024 2013

  65. [125]

    , author Stephen, S

    author Japkowicz, N. , author Stephen, S. , year 2002 . title The class imbalance problem: A systematic study1 . journal Intelligent Data Analysis volume 6 , pages 429--449 . https://journals.sagepub.com/action/showAbstract, :10.3233/IDA-2002-6504. note publisher: SAGE Publications

    work page doi:10.3233/ida-2002-6504 2002

  66. [126]

    Super-Resolved Canopy Height Mapping from Sentinel-2 Time Series Using Airborne LiDAR HD Reference Data across Metropolitan France

    author Kalinicheva, E. , author Helen, F. , author Mermoz, S. , author Mouret, F. , author Planells, M. , year 2025 . title Super-resolved canopy height mapping from Sentinel -2 time series using LiDAR HD reference data across metropolitan France . :10.48550/arXiv.2512.11524. note arXiv:2512.11524

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2512.11524 2025

  67. [127]

    , author Ozdemir, O

    author Karakutuk, A.K. , author Ozdemir, O. , author Senturk, S. , author Karakutuk, A.K. , author Ozdemir, O. , author Senturk, S. , year 2025 . title Optuna-optimized pythagorean fuzzy deep neural network: a novel framework for uncertainty-aware image classification . journal Applied Sciences volume 15 . :10.3390/app152011097

    work page doi:10.3390/app152011097 2025

  68. [128]

    On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima

    author Keskar, N.S. , author Mudigere, D. , author Nocedal, J. , author Smelyanskiy, M. , author Tang, P.T.P. , year 2017 . title On large-batcht training for deep learning: generalization gap and sharp minima . http://arxiv.org/abs/1609.04836, :10.48550/arXiv.1609.04836. note arXiv:1609.04836 [cs.LG]

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1609.04836 2017

  69. [129]

    , author Mohamed, A.H

    author Keskes, M.I. , author Mohamed, A.H. , author Borz, S.A. , author Niţă, M.D. , year 2025 . title Improving national forest mapping in Romania using machine learning and Sentinel -2 multispectral imagery . journal Remote Sensing volume 17 , pages 715 . https://www.mdpi.com/2072-4292/17/4/715, :10.3390/rs17040715

    work page doi:10.3390/rs17040715 2025

  70. [130]

    , author Fahey, R

    author LaRue, E.A. , author Fahey, R. , author Fuson, T.L. , author Foster, J.R. , author Matthes, J.H. , author Krause, K. , author Hardiman, B.S. , year 2022 . title Evaluating the sensitivity of forest structural diversity characterization to LiDAR point density . journal Ecosphere volume 13 , pages e4209 . https://esajournals.onlinelibrary.wiley.com/d...

    work page doi:10.1002/ecs2.4209 2022

  71. [131]

    , author Watt, M.S

    author Leonardo, E.M.C. , author Watt, M.S. , author Pearse, G.D. , author Dash, J.P. , author Persson, H.J. , year 2020 . title Comparison of TanDEM - X InSAR data and high-density ALS for the prediction of forest inventory attributes in plantation forests with steep terrain . journal Remote Sensing of Environment volume 246 , pages 111833 . https://www....

    work page doi:10.1016/j.rse.2020.111833 2020

  72. [132]

    , author Shen, X

    author Liu, H. , author Shen, X. , author Cao, L. , author Yun, T. , author Zhang, Z. , author Fu, X. , author Chen, X. , author Liu, F. , year 2021 . title Deep learning in forest structural parameter estimation using airborne LiDAR data . journal IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing volume 14 , pages 1603--161...

    work page doi:10.1109/jstars.2020.3046053 2021

  73. [133]

    , author Yang, J

    author Luo, B. , author Yang, J. , author Shi, S. , author Gan, R. , author Wu, Z. , author Wang, S. , author Wang, A. , author Du, L. , author Gong, W. , year 2026 . title InceptionFormer : A deep learning framework for UAV LiDAR point cloud completion to improve tree parameters estimation in dense forests . journal Remote Sensing of Environment volume 3...

    work page doi:10.1016/j.rse.2026.115348 2026

  74. [134]

    , author Packalen, P

    author Maltamo, M. , author Packalen, P. , author Laukkanen, L. , author Korhonen, L. , year 2025 . title The transferability and cross-use of airborne laser scanning-based leaf-off and leaf-on biomass models . journal European Journal of Remote Sensing volume 58 , pages 2542870 . https://doi.org/10.1080/22797254.2025.2542870, :10.1080/22797254.2025.25428...

    work page doi:10.1080/22797254.2025.2542870 2025

  75. [135]

    , author Hermosilla, T

    author Matasci, G. , author Hermosilla, T. , author Wulder, M.A. , author White, J.C. , author Coops, N.C. , author Hobart, G.W. , author Zald, H.S.J. , year 2018 . title Large-area mapping of Canadian boreal forest cover, height, biomass and other structural attributes using Landsat composites and lidar plots . journal Remote Sensing of Environment volum...

    work page doi:10.1016/j.rse.2017.12.020 2018

  76. [136]

    , author Breidenbach, J

    author Miettinen, J. , author Breidenbach, J. , author Adame, P. , author Adolt, R. , author Alberdi, I. , author Antropov, O. , author Arnarsson, O. , author Astrup, R. , author Berger, A. , author Bogason, J. , author Chirici, G. , author Corona, P. , author D'Amico, G. , author Fejfar, J. , author Fischer, C. , author Gohon, F. , author Gschwantner, T....

    work page doi:10.1016/j.dib.2025.111613 2025

  77. [137]

    , author Kane, V.R

    author Moran, C.J. , author Kane, V.R. , author Seielstad, C.A. , year 2020 . title Mapping forest canopy fuels in the western United States with LiDAR – Landsat covariance . journal Remote Sensing volume 12 . https://www.mdpi.com/2072-4292/12/6/1000, :10.3390/rs12061000. note publisher: Multidisciplinary Digital Publishing Institute

    work page doi:10.3390/rs12061000 2020

  78. [138]

    , author Gdulová, K

    author Moudrý, V. , author Gdulová, K. , author Fogl, M. , author Klápště, P. , author Urban, R. , author Komárek, J. , author Moudrá, L. , author Štroner, M. , author Barták, V. , author Solský, M. , year 2019 . title Comparison of leaf-off and leaf-on combined UAV imagery and airborne LiDAR for assessment of a post-mining site terrain and vegetation str...

    work page doi:10.1016/j.apgeog.2019.02.002 2019

  79. [139]

    , author Popescu, S.C

    author Narine, L.L. , author Popescu, S.C. , author Malambo, L. , year 2023 . title A methodological framework for mapping canopy cover using ICESat -2 in the southern USA . journal Remote Sensing volume 15 . https://www.mdpi.com/2072-4292/15/6/1548, :10.3390/rs15061548. note publisher: Multidisciplinary Digital Publishing Institute

    work page doi:10.3390/rs15061548 2023

  80. [140]

    , author Nordkvist, K

    author Nilsson, M. , author Nordkvist, K. , author Jonzén, J. , author Lindgren, N. , author Axensten, P. , author Wallerman, J. , author Egberth, M. , author Larsson, S. , author Nilsson, L. , author Eriksson, J. , author Olsson, H. , year 2017 . title A nationwide forest attribute map of Sweden predicted using airborne laser scanning data and field data...

    work page doi:10.1016/j.rse.2016.10.022 2017

Showing first 80 references.