Pith. sign in

REVIEW 3 major objections 4 minor 60 references

Archaeoscape: Bringing Aerial Laser Scanning Archaeology to the Deep Learning Era

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read This paper introduces Archaeoscape, a 888 km² open-access airborne laser scanning dataset of Angkorian Cambodia with 31,411 annotated archaeological features, and shows through benchmarks that deep learning models still struggle to detect…

desk verdict Large, well-documented ALS archaeology dataset with a credible benchmark; the label-completeness caveat is real but acknowledged and does not sink the main contribution. read the letter →

arxiv 2412.05203 v2 pith:6EM4MIW5 submitted 2024-12-06 cs.CV cs.AI

Yohann Perron , Vladyslav Sydorov , Adam P. Wijker , Damian Evans , Christophe Pottier , Loic Landrieu This is my paper · ORCID
classification cs.CVcs.AI
keywords airbornelaserscanningLiDARarchaeologysemanticsegmentationAngkorianCambodiaopen-accessdatasetbenchmarkdigitalterrainmodel
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 introduces Archaeoscape, an open-access airborne laser scanning (ALS) dataset covering 888 km² of Cambodia with 31,411 expert-drawn, field-verified archaeological features from the Angkorian period. The authors aim to establish that this is the largest open ALS archaeology dataset published so far and the first to release data, annotations, and models together, so that deep-learning methods can be trained and compared on a shared benchmark. Their benchmark of modern segmentation models shows the problem is not solved: the best model reaches roughly 53% mean overlap accuracy, elevation is the decisive input, and models pretrained on natural images transfer poorly. A fair reader would take the paper's contribution to be a hard, reusable benchmark that challenges computer vision to find subtle human-made structures hidden under dense jungle.

What carries the argument

The load-bearing object is the dataset itself, assembled as 23 non-overlapping parcels of 0.5 m resolution GeoTIFFs covering 888 km², each pixel carrying RGB orthophotography, a normalized Digital Terrain Model (nDTM) elevation, and a semantic label. The annotation scheme maps 12 expert feature types into five classes (temple, mound, hydrology, background, and a void class excluded from evaluation), and the benchmark treats the problem as pixel-wise semantic segmentation with class-wise intersection-over-union as the metric. The evaluation protocol adapts RGB-pretrained networks to the extra elevation channel by randomly initializing only the new first-layer weights, and it uses spatially distinct parcels with a 100 m buffer to prevent contamination, with the test set mixing remote parcels and parcels adjacent to training areas.

What would settle it

A second independent team of Khmer archaeologists re-draws all polygons on a 10 km² sample of the test parcels without seeing the released labels; if the two labelings disagree on many mound and hydrology boundaries, then the 1–2 point accuracy gaps between the top models are within annotation noise and the benchmark ranking is not settled.

Watch

Extended reading notes

Core claim

The paper's central claim is that Archaeoscape fills the missing resource for deep-learning archaeology: at 888 km² and 31,411 annotated instances, it is the largest open-access ALS archaeology dataset published to date and the first to distribute the imagery, the polygon annotations, and trained models openly. On its own benchmark, the authors find that modern semantic segmentation networks do not solve the problem. The best score, 52.8% mean intersection-over-union, comes from a U-Net at 512×512 input, while the best 224-level model is the hierarchical transformer PVTv2 at 52.1%; plain ViTs, CLIP, DINOv2, ScaleMAE, and DOFA all fare worse. The paper attributes the difficulty to the subtlety of earthen features, the need for kilometer-scale context, and the fact that the normalized terrain elevation, not RGB, carries most of the information. These findings lead the authors to argue that bespoke models and pretraining strategies are needed for ALS archaeology.

Load-bearing premise

The ground truth is the set of expert-drawn polygons, and the paper concedes that some real features in the background are probably missing; if the labels are incomplete or their boundaries are systematically wrong, all reported accuracy scores are biased and the ranking of models could change.

Editorial extensions

If this is right

  • Researchers can now compare semantic segmentation models on a common, open ALS archaeology benchmark with fixed train/validation/test parcels and released code, making results reproducible across groups.
  • RGB-only models are not viable for this task: the ablation shows they fall to about 30% mean intersection-over-union, so future systems must include elevation data.
  • The poor transfer of CLIP, DINOv2, ScaleMAE, and DOFA indicates that pretraining on natural or other satellite imagery does not substitute for data like Archaeoscape, and it argues for elevation-specific pretraining.
  • Because larger input windows improve detection of kilometer-scale hydrology features, architectures with large effective receptive fields are a promising direction, and the 512-pixel input size is a practical default.
  • The best reported performance, around 53% mean intersection-over-union, leaves substantial headroom, framing automatic ALS archaeology as an open problem rather than a solved application.

Reading between the lines

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

  • Beyond the paper's claims, the 'void' label concentrates ambiguous and excluded structures, so a model is never rewarded for predicting those regions; treating void as a fourth learnable class or modeling label uncertainty could alter the ranking of methods.
  • Because the dataset strips georeferencing and covers only Khmer sites, it should be read as a benchmark for recognizing Angkorian feature types, not as a general archaeological prospecting tool; cross-dataset tests on Maya or European ALS data would quantify how much transfers.
  • The test split mixes remote and adjacent parcels, which lets future work separate domain-shift failures from context-scale failures by reporting per-parcel scores; the authors describe this scenario but do not tabulate the breakdown.
  • Annotations accumulated over three decades, so label noise is likely structured by evolving survey methods; a bounded re-annotation study on a sample would show whether the small score differences between top models are meaningful.
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 / 4 minor

Summary. The paper introduces Archaeoscape, a LiDAR-derived archaeological dataset covering 888 km^2 in Cambodia with expert-annotated features (temples, mounds, hydrology, void/background), distributed as 0.5 m orthophotos and normalized digital terrain models. The authors benchmark CNN, ViT, and hierarchical ViT segmentation models, report mIoU values around 50, and analyze the importance of the elevation channel and input context size. They release data, annotations, and models under a credentialized open-access license and describe measures to prevent misuse. The central claims are that Archaeoscape is the largest open-access ALS archaeology dataset and the first such resource with open data, annotations, and models.

Significance. If the dataset and benchmark are sound, this is a substantial community resource: the area covered is several times larger than existing released ALS archaeology datasets, the annotation work is long-term and field-verified, and the split, preprocessing, and evaluation protocols are described in unusual detail. The benchmark is broad, covering modern architectures with various pretraining strategies, and the paper provides useful reproducibility information such as compute budgets, dataloader details, and hyperparameters. The main caveats are the unmeasured completeness of the expert labels, which affects all reported IoU scores, and the absence of error bars for the benchmark comparisons; both are fixable and do not negate the value of the released resource.

major comments (3)
  1. [Section 4.4 / Table 2] The annotation-completeness assumption is load-bearing. The paper states in Section 4.4 that 'we cannot rule out that background terrain may contain some yet uncovered features,' and the datasheet (Q7) claims only that the 'vast majority' of features have been identified, without quantifying recall. Since Background is the negative class in all IoU computations, any real feature missed by the annotators is scored as a false negative whenever a model detects it, and missed features are likely to be the subtlest ones, potentially correlated with terrain and class. The reported mIoU values and model rankings in Table 2 are therefore conditional on this unmeasured label recall. Please add an independent assessment of annotation completeness, for example a sample re-reviewed by independent experts or a per-class recall estimate, and discuss how the headline numbers would change under plausible miss rates.
  2. [Abstract, Section 1, Conclusion] The instance count is inconsistent across the manuscript: the Abstract says 31,141, Section 1 and Table 1 say 31,411, and the Conclusion says 31,144. Because the headline claim of being the largest open-access ALS archaeology dataset depends on this number, the discrepancy must be resolved and a single canonical count used consistently throughout.
  3. [Checklist A.1(2c) / Table 2] The paper reports no error bars, explicitly acknowledged in the checklist as being omitted due to computational constraints. Several benchmark conclusions rest on small margins, for example U-Net (50.5) versus HybViT (50.4) at 224 resolution, and PVTv2-b1 (52.1) versus SWIN-S (51.9). Without repeated-seed variance or another uncertainty measure, the relative ranking in Table 2 is not statistically supported. Please report standard deviations or confidence intervals for the main configurations, or otherwise justify that the observed differences exceed run-to-run variability.
minor comments (4)
  1. [Section 2.1 / Table 1] The sentence stating that Archaeoscape 'covers a 2 × larger surface area and contains 3× more instances than the next-largest closed archaeology LiDAR dataset' does not match Table 1 for either plausible comparator: relative to AHN-2 the area is about 2.0× but instances are about 8.2×, while relative to Chactún the area is about 3.9× but instances are about 2.9×. Please specify the intended comparator and correct the multipliers.
  2. [Table 4] In the RGB-only PVTv2-b1 row, the OA column shows 33.9, which equals the mIoU value and appears to be a typo; the background IoU is 82.0, so the overall accuracy should be much higher.
  3. [Table 2 and Section 4.2] There are several typographical errors in the table and surrounding text, including 'ImagineNet1K' for ImageNet1K, 'Webold' for 'We bold', 'present report performance' for 'we present the performance', and 'a least' for 'at least' in Section 3.1.
  4. [Section 2.2] The hybrid model 'HybViT' is described but no citation or architectural reference is provided; please cite the source or specify the exact configuration used in the experiments.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: benchmark scores are measured on expert ground truth; limitations are data-quality issues, not circular derivations.

full rationale

Archaeoscape is a dataset-and-benchmark paper without a derivation chain that reduces to its own inputs. The main empirical claims are that the dataset is the largest open-access ALS archaeology resource and that certain models achieve the reported IoU scores. The IoU scores follow from a standard supervised pipeline: expert-drawn polygons define ground truth, models are trained on the training split (Section 4.1) and evaluated on spatially disjoint test parcels with a 100 m buffer (Section 3.1), and OA/IoU/mIoU are standard definitions applied to test predictions. No parameter is fitted to the held-out labels and then reported as a prediction. The open-access claim rests on the paper's own footnote-1 definition of open access (data, annotations, and train/test split accessible), which is a definitional choice rather than a circular argument. Section 4.4 acknowledges a real benchmark-validity limitation: 'we cannot rule out that background terrain may contain some yet uncovered features.' This affects the trustworthiness of every reported IoU score if missed features are scored as Background, but it is an annotation-completeness issue, not circularity: the expert labels are external field-verified evidence, not outputs of the benchmark. Prior acquisition work by the authors is cited for raw LiDAR provenance, which is legitimate and not load-bearing self-citation. There is no uniqueness theorem, ansatz, or renamed known result. Verdict: no significant circularity; the benchmark's validity depends on the disclosed label-recall assumption.

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

The central claim rests on no fitted parameters; the dataset is a measurement and curation effort. The load-bearing assumptions concern label quality, spatial independence of splits, and the neutrality of the Void class exclusion. No new physical or conceptual entities are introduced.

assumptions (3)
  • domain assumption Expert-drawn, field-verified polygon annotations are accurate and complete enough to be used as ground truth for training and evaluation.
    All supervision and benchmark scores depend on these labels. Section 4.4 concedes that background may contain features that eluded detection, which could make the labels incomplete.
  • domain assumption A 100 meter buffer between parcels is sufficient to prevent spatial data leakage between train, validation, and test splits.
    Section 3.1 relies on this buffer to claim contamination-free splits, yet two test parcels are adjacent to training and validation parcels and terrain features can have extended spatial autocorrelation.
  • domain assumption Removing Void pixels from supervision and evaluation does not systematically bias the benchmark.
    Void covers 2.5% of pixels and represents ambiguous or excluded areas. If these areas align with hard examples, excluding them could inflate measured performance. The paper gives no sensitivity analysis.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Archaeoscape: Bringing Aerial Laser Scanning Archaeology to the Deep Learning Era." pith.science (2026). https://pith.science/paper/6EM4MIW5

@misc{pith2026241205203,
  author       = {Pith},
  title        = {Pith review of: Archaeoscape: Bringing Aerial Laser Scanning Archaeology to the Deep Learning Era},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6EM4MIW5}},
  note         = {Machine review of arXiv:2412.05203}
}
abstract

Airborne Laser Scanning (ALS) technology has transformed modern archaeology by unveiling hidden landscapes beneath dense vegetation. However, the lack of expert-annotated, open-access resources has hindered the analysis of ALS data using advanced deep learning techniques. We address this limitation with Archaeoscape (available at https://archaeoscape.ai/data/2024/), a novel large-scale archaeological ALS dataset spanning 888 km$^2$ in Cambodia with 31,141 annotated archaeological features from the Angkorian period. Archaeoscape is over four times larger than comparable datasets, and the first ALS archaeology resource with open-access data, annotations, and models. We benchmark several recent segmentation models to demonstrate the benefits of modern vision techniques for this problem and highlight the unique challenges of discovering subtle human-made structures under dense jungle canopies. By making Archaeoscape available in open access, we hope to bridge the gap between traditional archaeology and modern computer vision methods.

Figures

Figures reproduced from arXiv: 2412.05203 by the authors.

Figure 1
Figure 1. Archaeoscape. Our proposed dataset contains 888 km2 of aerial laser scans taken in Cambodia. The 3D point cloud LiDAR data (left) was processed to obtain a digital terrain model (middle). Archaeologists have drawn and field-verified 31,411 individual polygons by delineating anthropogenic features (right). (nDTM), encompassing over 3.5 billion pixels with RGB values, nDTM elevation, and semantic annotations. Traditio… view at source ↗
Figure 2
Figure 2. Archaeoscape overview. We show the vectorial annotations overlaid onto the relative elevation maps for each parcel, and their assignment to the training, validation or test splits. The position and orientation of the parcels is arbitrary. The geometry of the annotations has been simplified to reduce the file size of the paper. Best viewed on a computer screen. 3 Archaeoscape In this section, we describe the content … view at source ↗
Figure 3
Figure 3. Archaeoscape classes. We illustrate the three main classes with in-situ images (top row), top-view hillshaded elevation maps (middle row), and our annotations (bottom row). In many cases, the sought features are difficult to detect visually by in-situ observation but are more apparent on elevation maps. use cases: predicting features in a new area under a domain shift, and a realistic scenario in which archaeologist… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Qualitative performance. We provide examples of input elevation maps (a) and their corresponding annotations (b), as well as the prediction of a standard U-Net (c) and our best model (d)—improvements in green. The red squares represent the size of the input images: 224…
Figure 5
Figure 5. Figure 5: Channel ablation. We represent the orthophotography (a), normalized terrain model (b), and annotations (c). We also provide the prediction of a PVTv2 model operating on RGB photos (d), and a model processing both RGB and elevation data (e). The model using only radiome…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 55 canonical work pages

  1. [1]

    Geospatial revolution and remote sensing LiDAR in Mesoamerican archaeology

    Arlen F Chase, Diane Z Chase, Christopher T Fisher, Stephen J Leisz, and John F Weishampel. Geospatial revolution and remote sensing LiDAR in Mesoamerican archaeology. Proceedings of the National Academy of Sciences, 2012

    work page 2012

  2. [2]

    Progres- sion and issues in the Mesoamerican geospatial revolution: An introduction

    Arlen F Chase, Kathryn Reese-Taylor, Juan C Fernandez-Diaz, and Diane Z Chase. Progres- sion and issues in the Mesoamerican geospatial revolution: An introduction. Advances in Archaeological Practice, 2016

    work page 2016

  3. [3]

    Uncovering archaeological landscapes at Angkor using LiDAR

    Damian Evans, Roland J Fletcher, Christophe Pottier, Jean-Baptiste Chevance, Dominique Soutif, Boun Suy Tan, Sokrithy Im, Darith Ea, Tina Tin, Samnang Kim, et al. Uncovering archaeological landscapes at Angkor using LiDAR. Proceedings of the National Academy of Sciences, 2013

    work page 2013

  4. [4]

    Ancient lowland Maya complexity as revealed by airborne laser scanning of northern Guatemala

    Marcello A Canuto, Francisco Estrada-Belli, Thomas G Garrison, Stephen D Houston, Mary Jane Acuña, Milan Ková ˇc, Damien Marken, Philippe Nondédéo, Luke Auld-Thomas, Cyril Castanet, et al. Ancient lowland Maya complexity as revealed by airborne laser scanning of northern Guatemala. Science, 2018

    work page 2018

  5. [5]

    Combined detection and segmentation of archeological structures from LiDAR data using a deep learning approach

    Alexandre Guyot, Marc Lennon, Thierry Lorho, and Laurence Hubert-Moy. Combined detection and segmentation of archeological structures from LiDAR data using a deep learning approach. Journal of Computer Applications in Archaeology, 2021

    work page 2021

  6. [6]

    Semantic segmentation (U-Net) of archaeological features in airborne laser scanning—example of the Białowie˙za forest

    Paweł Zbigniew Banasiak, Piotr Leszek Berezowski, Rafał Zapłata, Miłosz Mielcarek, Konrad Duraj, and Krzysztof Stere´nczak. Semantic segmentation (U-Net) of archaeological features in airborne laser scanning—example of the Białowie˙za forest. Remote Sensing, 2022

    work page 2022

  7. [7]

    Ecology in an anthropogenic biosphere

    Erle C Ellis. Ecology in an anthropogenic biosphere. Ecological Monographs, 2015

    work page 2015

  8. [8]

    LiDAR as a tool for archaeological research: A case study

    James Schindling and Cerian Gibbes. LiDAR as a tool for archaeological research: A case study. Archaeological and Anthropological Sciences, 2014

    work page 2014

Show all 60 references
  1. [9]

    U-Net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-Net: Convolutional networks for biomedical image segmentation. In MICCAI. Springer, 2015

    2015

  2. [10]

    LiDAR applica- tions in archaeology: A systematic review

    Giacomo Vinci, Federica Vanzani, Alessandro Fontana, and Stefano Campana. LiDAR applica- tions in archaeology: A systematic review. Archaeological Prospection, 2024

    2024

  3. [11]

    Looting hoards of gold and poaching spotted owls: Data confidentiality among archaeologists & zoologists

    Rebecca D Frank, Adam Kriesberg, Elizabeth Yakel, and Ixchel M Faniel. Looting hoards of gold and poaching spotted owls: Data confidentiality among archaeologists & zoologists. Proceedings of the Association for Information Science and Technology, 2015

    2015

  4. [12]

    Ethics in archaeological lidar

    Anna Cohen, Sarah Klassen, and Damian Evans. Ethics in archaeological lidar. Journal of Computer Applications in Archaeology, 2020

    2020

  5. [13]

    https://github.com/ickramer/Arran

    Arran: A benchmark dataset for automated object detection of archaeological sites on LiDAR data. https://github.com/ickramer/Arran. Accessed: 2024-02-29

    2024

  6. [14]

    Johnson, and Chandi Witharana

    Ji Won Suh, Eli Anderson, William Ouimet, Katharine M. Johnson, and Chandi Witharana. Mapping relict charcoal hearths in new england using deep convolutional neural networks and 11 LiDAR data. Remote Sensing, 2021

    2021

  7. [15]

    Jincheng Zhang, William Ringle, and Andrew R. Willis. Unveiling ancient Maya settlements using aerial LiDAR image segmentation. arXiv preprint arXiv:2403.05773, 2024

    2024 arXiv

  8. [16]

    Bourgeois

    Wouter Verschoof-van der Vaart, Karsten Lambers, Wojtek Kowalczyk, and Quentin P.J. Bourgeois. Combining deep learning and location-based ranking for large-scale archaeo- logical prospection of LiDAR data from the netherlands. ISPRS International Journal of Geo-Information, 2020

    2020

  9. [17]

    Deep learning for archaeological object detection on LiDAR: New evaluation measures and insights

    Marco Fiorucci, Wouter Verschoof-van der Vaart, Paolo Soleni, Bertrand Le Saux, and Arianna Traviglia. Deep learning for archaeological object detection on LiDAR: New evaluation measures and insights. Remote Sensing, 2022

    2022

  10. [18]

    Automated large-scale mapping and analysis of relict charcoal hearths in Connecticut (USA) using a deep learning YOLOv4 framework

    Wouter Verschoof-van der Vaart, Alexander Bonhage, Anna Schneider, William Ouimet, and Thomas Raab. Automated large-scale mapping and analysis of relict charcoal hearths in Connecticut (USA) using a deep learning YOLOv4 framework. Archaeological Prospection, 2023

    2023

  11. [19]

    Bringing lunar LiDAR back down to Earth: Mapping our industrial heritage through deep transfer learning

    Jane Gallwey, Matthew Eyre, Matthew Tonkins, and John Coggan. Bringing lunar LiDAR back down to Earth: Mapping our industrial heritage through deep transfer learning. Remote Sensing, 2019

    2019

  12. [20]

    Carter, Jeff H

    Benjamin P. Carter, Jeff H. Blackadar, and Weston L. A. Conner. When computers dream of charcoal: Using deep learning, open tools, and open data to identify relict charcoal hearths in and around state game lands in Pennsylvania. Advances in Archaeological Practice, 9, 2021. do...

    2021 doi

  13. [21]

    Semantic segmentation of airborne LiDAR data in Maya archaeology

    Marek Bundzel, Miroslav Jašˇcur, Milan Kováˇc, Tibor Lieskovský, Peter Sinˇcák, and Tomáš Tkáˇcik. Semantic segmentation of airborne LiDAR data in Maya archaeology. Remote Sensing, 2020

    2020

  14. [22]

    Learning to classify structures in ALS-derived visualizations of ancient Maya settlements with CNN

    Maja Somrak, Sašo Džeroski, and Žiga Kokalj. Learning to classify structures in ALS-derived visualizations of ancient Maya settlements with CNN. Remote Sensing, 2020

    2020

  15. [23]

    Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position

    Kunihiko Fukushima. Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological cybernetics, 1980

    1980

  16. [24]

    Gradient-based learning applied to document recognition

    Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 1998

    1998

  17. [25]

    Rethinking atrous convolution for semantic image segmentation

    Liang-Chieh Chen, George Papandreou, Florian Schroff, and Hartwig Adam. Rethinking atrous convolution for semantic image segmentation. arXiv preprint arXiv:1706.05587, 2017

    2017 arXiv

  18. [26]

    Spatial pyramid pooling in deep convolutional networks for visual recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Spatial pyramid pooling in deep convolutional networks for visual recognition. TPAMI, 2015

    2015

  19. [27]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. NeurIPS, 2017

    2017

  20. [28]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. An image is worth 16x16 words: Transformers for image recognition at scale. ICLR, 2021

    2021

  21. [29]

    Neural plasticity-inspired foundation model for observing the Earth crossing modalities

    Zhitong Xiong, Yi Wang, Fahong Zhang, Adam J Stewart, Joëlle Hanna, Damian Borth, Ioannis Papoutsis, Bertrand Le Saux, Gustau Camps-Valls, and Xiao Xiang Zhu. Neural plasticity-inspired foundation model for observing the Earth crossing modalities. arXiv preprint arXiv:2403.15356, 2024

    2024

  22. [30]

    Pyramid vision transformer: A versatile backbone for dense prediction without convolutions

    Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, and Ling Shao. Pyramid vision transformer: A versatile backbone for dense prediction without convolutions. In ICCV, 2021. 12

    2021

  23. [31]

    SWIN transformer: Hierarchical vision transformer using shifted windows

    Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo. SWIN transformer: Hierarchical vision transformer using shifted windows. In ICCV, 2021

    2021

  24. [32]

    Twins: Revisiting the design of spatial attention in vision transformers

    Xiangxiang Chu, Zhi Tian, Yuqing Wang, Bo Zhang, Haibing Ren, Xiaolin Wei, Huaxia Xia, and Chunhua Shen. Twins: Revisiting the design of spatial attention in vision transformers. NeurIPS, 2021

    2021

  25. [33]

    PVTv2: Improved baselines with pyramid vision transformer

    Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, and Ling Shao. PVTv2: Improved baselines with pyramid vision transformer. Computational Visual Media, 2022

    2022

  26. [34]

    Learning transferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervision. In ICML, 2021

    2021

  27. [35]

    LAION-5B: An open large-scale dataset for training next generation image-text models

    Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon, Ross Wightman, Mehdi Cherti, Theo Coombes, Aarush Katta, Clayton Mullis, Mitchell Wortsman, Patrick Schramowski, Srivatsa Kundurthy, Katherine Crowson, Ludwig Schmidt, Robert Kaczmarczyk, and Jenia Jitsev. LAION...

    2022

  28. [36]

    DINOv2: Learning robust visual features without supervision

    Maxime Oquab, Timothée Darcet, Théo Moutakanni, Huy V V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel HAZIZA, Francisco Massa, Alaaeldin El-Nouby, et al. DINOv2: Learning robust visual features without supervision. TMLR, 2023

    2023

  29. [37]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. Masked autoencoders are scalable vision learners. In CVPR, 2022

    2022

  30. [38]

    Scale-MAE: A scale- aware masked autoencoder for multiscale geospatial representation learning

    Colorado J Reed, Ritwik Gupta, Shufan Li, Sarah Brockman, Christopher Funk, Brian Clipp, Kurt Keutzer, Salvatore Candido, Matt Uyttendaele, and Trevor Darrell. Scale-MAE: A scale- aware masked autoencoder for multiscale geospatial representation learning. In ICCV, 2023

    2023

  31. [39]

    Airborne laser scanning as a method for exploring long-term socio-ecological dynamics in Cambodia

    Damian Evans. Airborne laser scanning as a method for exploring long-term socio-ecological dynamics in Cambodia. Journal of Archaeological Science, 2016

    2016

  32. [40]

    Terrasolid

    Terrasolid Ltd. Terrasolid. https://terrasolid.com/, 2024. [Online; accessed 12-May- 2024]

    2024

  33. [41]

    Automatic extraction of irregular network digital terrain models

    Robert J Fowler and James J Little. Automatic extraction of irregular network digital terrain models. In Conference on Computer graphics and interactive techniques, 1979

    1979

  34. [42]

    Carte archéologique de la région d’Angkor - Zone Sud

    Christophe Pottier. Carte archéologique de la région d’Angkor - Zone Sud . PhD thesis, Université de la Sorbonne Nouvelle - Paris 3, 1999

    1999

  35. [43]

    hydraulic city

    Damian Evans. Putting Angkor on the map: A new survey of a Khmer “hydraulic city” in historical and theoretical context. PhD thesis, University of Sydney, 2007

    2007

  36. [44]

    ImageNet large scale visual recognition challenge

    Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, et al. ImageNet large scale visual recognition challenge. IJCV, 2015

    2015

  37. [45]

    ImageNet-21K pretraining for the masses

    Tal Ridnik, Emanuel Ben-Baruch, Asaf Noy, and Lihi Zelnik-Manor. ImageNet-21K pretraining for the masses. In NeurIPS Datasets and Benchmarks Track, 2021

    2021

  38. [46]

    SatlasPretrain: A large-scale dataset for remote sensing image understanding

    Favyen Bastani, Piper Wolters, Ritwik Gupta, Joe Ferdinando, and Aniruddha Kembhavi. SatlasPretrain: A large-scale dataset for remote sensing image understanding. In ICCV, 2023

    2023

  39. [47]

    Karsten Lambers, Wouter Verschoof-van der Vaart, and Quentin P. J. Bourgeois. Integrating remote sensing, machine learning, and citizen science in Dutch archaeological prospection. Remote Sensing, 2019. 13

    2019

  40. [48]

    Discover the mysteries of the Maya: Selected contributions from the machine learning challenge & the discovery challenge workshop at ECML PKDD 2021, 2022

    Dragi Kocev, Nikola Simidjievski, Ana Kostovska, Ivica Dimitrovski, and Žiga Kokalj. Discover the mysteries of the Maya: Selected contributions from the machine learning challenge & the discovery challenge workshop at ECML PKDD 2021, 2022

    2021

  41. [49]

    Davis and Julius Lundin

    Dylan S. Davis and Julius Lundin. Locating charcoal production sites in Sweden using LiDAR, hydrological algorithms, and deep learning. Remote Sensing, 2021

    2021

  42. [50]

    Orengo, Felipe Lumbreras, Miguel Carrero-Pazos, João Fonte, and Benito Vilas-Estévez

    Iban Berganzo-Besga, Hector A. Orengo, Felipe Lumbreras, Miguel Carrero-Pazos, João Fonte, and Benito Vilas-Estévez. Hybrid MSRM-based deep learning and multitemporal Sentinel 2-based machine learning algorithm detects near 10k archaeological Tumuli in north-western Iberia. Re...

    2021

  43. [51]

    Detecting archaeological features with airborne laser scanning in the alpine tundra of Sápmi, northern Finland

    Oula Seitsonen and Janne Ikäheimo. Detecting archaeological features with airborne laser scanning in the alpine tundra of Sápmi, northern Finland. Remote Sensing, 2021

    2021

  44. [52]

    Richards-Rissetto, D

    H. Richards-Rissetto, D. Newton, and A. Al Zadjali. A 3D point cloud deep learning approach using LiDAR to identify ancient Maya archaeological sites. ISPRS Annals of the Photogramme- try, Remote Sensing and Spatial Information Sciences, 2021

    2021

  45. [53]

    Alexander Bonhage, Mahmoud Eltaher, Thomas Raab, Michael Breuß, Alexandra Raab, and Anna Schneider. A modified mask region-based convolutional neural network approach for the automated detection of archaeological sites on high-resolution light detection and ranging- derived di...

    2021

  46. [54]

    LoRa: Low-rank adaptation of large language models

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. LoRa: Low-rank adaptation of large language models. ICLR, 2022

    2022

  47. [55]

    org/docs/stable/generated/torch.optim

    PyTorch: ReduceLROnPlateau. org/docs/stable/generated/torch.optim. lr_scheduler.ReduceLROnPlateau.html#torch.optim.lr_scheduler. ReduceLROnPlateau. Accessed: 2024-02-29. 14 APPENDIX A.1 Checklist

    2024

  48. [56]

    For all authors... (a) Do the main claims made in the abstract and introduction accurately reflect the paper’s contributions and scope? [Yes] (b) Did you describe the limitations of your work? [Yes] See Section 4.4. (c) Did you discuss any potential negative societal impacts o...

  49. [57]

    for benchmarks)

    If you ran experiments (e.g. for benchmarks)... (a) Did you include the code, data, and instructions needed to reproduce the main experi- mental results (either in the supplemental material or as a URL)? [Yes] See supplemen- tary material. (b) Did you specify all the training ...

  50. [58]

    (a) If your work uses existing assets, did you cite the creators? [Yes] (b) Did you mention the license of the assets? [N/A] The code licenses are given on the linked websites

    If you are using existing assets (e.g., code, data, models) or curating/releasing new assets... (a) If your work uses existing assets, did you cite the creators? [Yes] (b) Did you mention the license of the assets? [N/A] The code licenses are given on the linked websites. (c) ...

  51. [59]

    DATASET”) available for research and educational purposes to individuals or entities (

    If you used crowdsourcing or conducted research with human subjects... (a) Did you include the full text of instructions given to participants and screenshots, if applicable? [N/A] (b) Did you describe any potential participant risks, with links to Institutional Review Board (...

  52. [2024]

    raw” data saved in addition to the preprocessed/cleaned/labeled data (e.g., to support unanticipated future uses)? If so, please provide a link or other access point to the “raw

    All persons involved were employees and researchers from foreign governmental institutions, such as the EFEO or Sydney University, or employed by the Cambo- dian governmental authorities, following strictly existing ethical codes and national regulations. Q25 Over what timefra...

    2012

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.