Pith. sign in

REVIEW 3 major objections 5 minor 30 references

Leveraging Satellite Image Time Series for Accurate Extreme Event Detection

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper argues that using several pre-disaster satellite images filters out irrelevant changes and isolates disaster signals, yielding more accurate extreme-event detection than before/after pairs.

desk verdict SITS-Extreme is a clean, plausible approach to multi-temporal extreme-event detection with solid gains on RaVÆn, but split ambiguity and test-set threshold selection mean the headline numbers should be read with caution. read the letter →

arxiv 2506.11544 v1 pith:YKYJEMXW submitted 2025-06-13 cs.CV

Heng Fang , Hossein Azizpour This is my paper · ORCID
classification cs.CV
keywords extremeeventdetectionsatelliteimagetimeserieschangecontrastivelearningconsistentlossautoencoderdisastermonitoringSentinel-2
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 aims to establish that satellite image time series, formed from several pre-disaster images and one post-disaster image, detect extreme events such as floods, fires, and landslides more accurately than bi-temporal before/after pairs. The proposed framework, SITS-Extreme, learns latent representations with a weight-shared transformer autoencoder and two task-specific losses, then decides whether an event occurred by thresholding the mean cosine distance between pre- and post-disaster representations. On the real-world RaVÆn dataset, the best variant reaches an Average Precision of 0.9069 and an F1 score of 0.8473, beating the strongest adapted baseline by 10.66% AP. The authors argue this matters because it offers a scalable way to monitor disasters using freely available low-to-medium resolution satellite imagery.

What carries the argument

The load-bearing mechanism is the Mean Cosine Distance (MCD) detector applied to a learned latent space. The latent space is produced by a weight-shared transformer-based autoencoder trained with reconstruction, contrastive, and consistency losses; the contrastive loss acts on disaster-affected series and the consistency loss on unaffected series. MCD averages the cosine distance between the post-disaster representation $h_T$ and each pre-disaster representation $h_i$, and a threshold $\tau$ selected on validation data converts the score into a binary prediction. The losses are what make cosine distance separate abrupt, disaster-induced change from recurring seasonal and atmospheric variation.

What would settle it

Re-run the RaVÆn evaluation with a strict area-based split, placing all patches from each Area of Interest into either training or testing, and compare the Average Precision; a large drop from the reported 0.9069 would show that the result depends on memorization of scene-specific appearance rather than a general extreme-event detector.

Watch

Extended reading notes

Core claim

The central discovery is that a simple thresholded mean cosine distance between pre-disaster and post-disaster latent representations becomes a reliable extreme-event indicator when the representations are learned with the right multi-temporal objective. The training objective combines a reconstruction loss, a contrastive loss for affected time series that pulls all pre-disaster representations together and pushes the post-disaster representation away from them, and a consistency loss for unaffected series that keeps all timesteps close. At inference, the Mean Cosine Distance (MCD) between the last image and each earlier image is averaged, and a validation-tuned threshold $\tau$ turns the score into a binary detection. The paper reports that this approach outperforms bi-temporal and multi-temporal baselines on both a synthetic dataset and the real-world RaVÆn dataset, with the largest margin on real data.

Load-bearing premise

The reported accuracy assumes that the 7:1:2 train/validation/test split keeps patches from the same disaster area out of both training and testing; if the split is not by Area of Interest, the test scores may overstate how well the model generalizes to new locations.

Editorial extensions

If this is right

  • Multi-temporal satellite imagery becomes a practical basis for early disaster response, since the framework runs on openly available Sentinel-2 data at 10–60 m resolution.
  • Increasing the number of pre-disaster timesteps improves detection: the paper's own comparison shows a five-step baseline beats its two-step counterpart by 8.08% AP, and the proposed method exploits the same temporal context.
  • Performance varies by disaster type: the model reaches AP above 0.93 for fires and hurricanes but drops to 0.7353 for floods and 0.4395 for landslides, indicating underrepresented classes need more data or different weighting.
  • The framework is resolution-agnostic, so the same representation and MCD setup can be adapted to very high-resolution imagery when available.

Reading between the lines

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

  • Our reading suggests the reported accuracy could be inflated by leakage: the paper describes a 7:1:2 split on RaVÆn without stating that it is by Area of Interest, so patches from the same AOI may appear in both training and testing, letting the model memorize scene-specific appearance rather than learn a general detector.
  • We infer that a single global threshold may not be optimal across disaster types, since fires and floods likely occupy different regions of the MCD distribution; per-disaster or per-region thresholds are a natural extension.
  • The contrastive-consistency objective could transfer to other abrupt-change detection problems, such as building damage assessment or deforestation monitoring, because it does not depend on the input modality.
  • A testable extension is to feed the same framework with SAR time series or fused optical-SAR data; the losses and MCD computation are modality-agnostic, so all-weather disaster detection is a plausible next step.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. This paper proposes SITS-Extreme, a framework that detects extreme events from satellite image time series by learning patch representations with a weight-shared transformer autoencoder. The training objective combines a reconstruction loss with a contrastive loss for affected series (pushing the post-disaster representation away from pre-disaster ones and pulling pre-disaster representations together) and a consistency loss for unaffected series. At inference, disaster is declared when the mean cosine distance between the post-disaster representation and the pre-disaster representations exceeds a threshold. Experiments on a synthetic EuroSAT-based dataset and the real RaVÆn dataset report large AP/F1 improvements over adapted SiamConcat/SiamDiff baselines, with ablations supporting the contribution of each loss term.

Significance. If the AOI-level generalization holds, the paper addresses a relevant and underexplored task, and the central idea—using multiple pre-disaster observations to filter irrelevant change—is well motivated. Strengths include the simple and interpretable inference rule, the component ablations in Table 3, and the evaluation across four disaster types. I verified that Eq. (2), as printed, is consistent with the stated objective: minimizing the negative log ratio pushes the post-disaster representation away from pre-disaster representations while pulling pre-disaster representations together, so the reported gains are not explained by a sign error. The main uncertainties are in the evaluation protocol rather than in the method's derivation.

major comments (3)
  1. [4.1] Section 4.1 says the preprocessed RaVÆn time series were divided into training, validation, and testing sets in a 7:1:2 ratio but does not state that the split is performed by Area of Interest. Since patches from the same AOI share geography, seasonality, and acquisition conditions, a random patch-level split can leak AOI-specific appearance into the training set and inflate the reported AP. Please state the split unit explicitly; if it is not AOI-disjoint, re-run the main comparison (Table 2) with an AOI-disjoint split and report AP/F1 for both settings.
  2. [4.3 / Appendix B] The F1, precision, and recall numbers in Tables 2-4 are computed with the threshold selected by grid search on the testing set. This maximizes the reported F1 on the test data and is not a valid estimate of deployable performance; it also makes F1 comparisons between methods potentially biased because the threshold adapts to each model's score distribution. Select the threshold on the validation set (or via nested cross-validation), and report both validation-selected and threshold-independent metrics.
  3. [4.2] The baseline set is limited to SiamConcat and SiamDiff variants. These are simple feature-interaction change detectors, and the claim of 'substantial improvements over widely used strong bi-temporal baselines' is not fully supported without at least one modern transformer-based change detector (e.g., ChangeFormer, BiT) or a remote-sensing foundation-model encoder adapted to the same multi-temporal setting. Please add such a comparison or soften the claim.
minor comments (5)
  1. [Eq. (5)] Equation (5) uses an absolute value in the cosine distance; this equates a patch with its negative, which may be undesirable for change detection. Please justify or remove the absolute value.
  2. [Eq. (2)] Equation (2) uses exp(D(·,·)) with D a distance; consider using a similarity kernel such as exp(-D) or 1-D to match the usual contrastive formulation and to avoid gradient saturation.
  3. [Abstract] The abstract says 'early detection,' but the protocol detects events at the final time step after they have occurred; consider 'rapid detection' or clarify the temporal setup.
  4. [Table 3] The sentence 'adding lconsist leads to a greater improvement compared to lcontra' depends on the order of addition; please state the comparison explicitly (e.g., AE+lcontra versus AE+lcontra+lconsist).
  5. [Reproducibility] The paper does not mention code release; providing the implementation would substantially improve reproducibility of the reported AP/F1 numbers.

Circularity Check

1 steps flagged · score 2.0 of 10

No load-bearing circularity: the central result is supervised learning benchmarked against external baselines, with AP as a threshold-independent metric. The only mild circularity is the test-set-fitted F1 threshold.

  1. fitted input called prediction [Section 4.3, Evaluation Metrics; Appendix B, Implementation Details]
    "F1, precision, and recall are determined by selecting the optimal threshold via grid search on the testing set to show the best possible performance."

    The reported F1/precision/recall values are computed on the same testing set used to select the threshold, so those numbers are in-sample optimized fits rather than predictions at a pre-specified operating point. By construction, the reported F1 is the maximum over thresholds on that test set and thus overstates expected operating-point performance. This is a minor, non-load-bearing circularity: the paper's primary claim rests on threshold-independent AP, the test-set threshold is explicitly labeled as demonstration-only, and the central comparison against external baselines does not reduce to this fit.

full rationale

The derivation chain is not circular in its load-bearing part. SITS-Extreme is a supervised representation-learning method whose detection score is the Mean Cosine Distance (Eq. 6), evaluated on held-out labels with AP as the threshold-independent primary metric and benchmarked against reimplemented external bi-temporal baselines (Table 2). No central claim is obtained by fitting a parameter and renaming it a prediction; the contrastive and consistency losses are stated training objectives, and the reported gains are empirical. The only mild circularity is the F1/precision/recall threshold selected on the testing set, which inflates those operating-point numbers but does not affect the AP-based core claim. Two caveats are correctness or validity concerns rather than circularity: the RaVÆn split is described only as a 7:1:2 ratio without stating an AOI-stratified split, creating a potential leakage risk, and Eq. 2 as printed appears to minimize the pre/post distance that the text says it maximizes, which is an internal-consistency and reproducibility defect rather than a reduction of the result to its inputs. Self-citations [14, 28] are contextual and not load-bearing, and no uniqueness theorem or ansatz is smuggled in through self-citation.

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

The framework does not postulate new physical entities. The empirical results depend on a handful of hyperparameters chosen on the validation set, plus the assumption that the dataset splits are leakage-free and the synthetic transformations are representative.

free parameters (3)
  • lambda (contrastive loss weight) = 0.5 (synthetic), 0.25 (RaVÆn)
    Chosen by grid search on validation AP; controls contribution of the contrastive loss.
  • mu (consistency loss weight) = 0.5
    Chosen by grid search on validation AP.
  • MCD threshold tau = not reported (selected on validation)
    Inference threshold in Eq. (7); selection on validation is standard but the specific value is not disclosed.
assumptions (3)
  • domain assumption The RaVÆn patch-level binary labels derived from the change masks are accurate enough to supervise the contrastive and consistency losses.
    Section 4.1 and Appendix C acknowledge that many pixel-level annotations are coarse and some patches are mislabeled; the method relies on these labels for training.
  • domain assumption Patches from the same AOI are independent after the 7:1:2 random split; no spatial autocorrelation leakage between train and test.
    Section 4.1 describes a random split without AOI separation; this independence assumption is required for the reported AP to reflect generalization.
  • domain assumption The synthetic EuroSAT transformations (ColorJitter, elliptical transparent masks, CutMix with Gaussian filter) adequately mimic the confounding seasonal/cloud changes and disaster signatures in real data.
    Section 4.1 and Algorithm 1; the synthetic dataset is used to validate the core assumption that disaster-relevant signals can be separated from recurring changes.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Leveraging Satellite Image Time Series for Accurate Extreme Event Detection." pith.science (2026). https://pith.science/paper/YKYJEMXW

@misc{pith2026250611544,
  author       = {Pith},
  title        = {Pith review of: Leveraging Satellite Image Time Series for Accurate Extreme Event Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YKYJEMXW}},
  note         = {Machine review of arXiv:2506.11544}
}
read the original abstract

Climate change is leading to an increase in extreme weather events, causing significant environmental damage and loss of life. Early detection of such events is essential for improving disaster response. In this work, we propose SITS-Extreme, a novel framework that leverages satellite image time series to detect extreme events by incorporating multiple pre-disaster observations. This approach effectively filters out irrelevant changes while isolating disaster-relevant signals, enabling more accurate detection. Extensive experiments on both real-world and synthetic datasets validate the effectiveness of SITS-Extreme, demonstrating substantial improvements over widely used strong bi-temporal baselines. Additionally, we examine the impact of incorporating more timesteps, analyze the contribution of key components in our framework, and evaluate its performance across different disaster types, offering valuable insights into its scalability and applicability for large-scale disaster monitoring.

Figures

Figures reproduced from arXiv: 2506.11544 by the authors.

Figure 1
Figure 1. Real-world application of the SITS-Extreme model for extreme event detection. Satellite observations are collected over time, [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of our framework SITS-Extreme. In the training stage, the left part illustrates the autoencoding setup, which learns [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. An example time series of the generated synthetic dataset. The base image (leftmost) is from the EuroSAT dataset [ [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: A scene from the real-world RaVÆn dataset illustrates a satellite patch time series affected by flood. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Histogram of patch change ratios, where the change ratio is defined as the proportion of changed pixels relative to the total pixels in each patch. The binary distribution shows that most patches are either entirely unchanged (change ratio close to 0) or fully affected…
Figure 6
Figure 6. Figure 6: Examples of pixel-level annotations in the dataset. Good annotations [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 19 canonical work pages

  1. [1]

    A transformer-based siamese network for change detection

    Wele Gedara Chaminda Bandara and Vishal M Patel. A transformer-based siamese network for change detection. In IGARSS 2022-2022 IEEE International Geoscience and Re- mote Sensing Symposium, pages 207–210. IEEE, 2022. 2

    work page 2022

  2. [2]

    End-to- end object detection with transformers

    Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, and Sergey Zagoruyko. End-to- end object detection with transformers. In European confer- ence on computer vision, pages 213–229. Springer, 2020. 4, 11

    work page 2020

  3. [3]

    Remote sensing im- age change detection with transformers

    Hao Chen, Zipeng Qi, and Zhenwei Shi. Remote sensing im- age change detection with transformers. IEEE Transactions on Geoscience and Remote Sensing, 60:1–14, 2021. 1, 2

    work page 2021

  4. [4]

    Semantic decoupled representation learning for re- mote sensing image change detection

    Hao Chen, Yifan Zao, Liqin Liu, Song Chen, and Zhen- wei Shi. Semantic decoupled representation learning for re- mote sensing image change detection. InIGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Sympo- sium, pages 1051–1054. IEEE, 2022. 1, 2

    work page 2022

  5. [5]

    A simple framework for contrastive learning of visual representations

    Ting Chen, Simon Kornblith, Mohammad Norouzi, and Ge- offrey Hinton. A simple framework for contrastive learning of visual representations. In International conference on ma- chine learning, pages 1597–1607. PMLR, 2020. 4

    2020

  6. [6]

    Satmae: Pre-training transformers for tem- poral and multi-spectral satellite imagery

    Yezhen Cong, Samar Khanna, Chenlin Meng, Patrick Liu, Erik Rozi, Yutong He, Marshall Burke, David Lobell, and Stefano Ermon. Satmae: Pre-training transformers for tem- poral and multi-spectral satellite imagery. Advances in Neu- ral Information Processing Systems , 35:197–211, 2022. 3, 4

    work page 2022

  7. [7]

    A change detection reality check

    Isaac Corley, Caleb Robinson, and Anthony Ortiz. A change detection reality check. arXiv preprint arXiv:2402.06994 ,

    arXiv

  8. [8]

    Centre for Research on the Epidemiology of Disasters (CRED) and US Agency Int. Dev. (USAID). Disaster year in review 2023., 2024. 1

    work page 2023

Show all 30 references
  1. [9]

    Fully convolutional siamese networks for change detection

    Rodrigo Caye Daudt, Bertr Le Saux, and Alexandre Boulch. Fully convolutional siamese networks for change detection. In 2018 25th IEEE international conference on image pro- cessing (ICIP), pages 4063–4067. IEEE, 2018. 2, 6

    2018

  2. [10]

    An image is worth 16x16 words: Trans- formers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, et al. An image is worth 16x16 words: Trans- formers for image recognition at scale. arXiv preprint a...

    2010 arXiv

  3. [11]

    Changer: Feature inter- action is what you need for change detection

    Sheng Fang, Kaiyu Li, and Zhe Li. Changer: Feature inter- action is what you need for change detection. IEEE Trans- actions on Geoscience and Remote Sensing , 61:1–11, 2023. 6

    2023

  4. [12]

    A simple, strong baseline for building damage detection on the xbd dataset

    Sebastian Gerard, Paul Borne-Pons, and Josephine Sullivan. A simple, strong baseline for building damage detection on the xbd dataset. arXiv preprint arXiv:2401.17271, 2024. 1

    2024 arXiv

  5. [13]

    Creating xbd: A dataset for assessing building damage from satellite imagery

    Ritwik Gupta, Bryce Goodman, Nirav Patel, Ricky Hosfelt, Sandra Sajeev, Eric Heim, Jigar Doshi, Keane Lucas, Howie Choset, and Matthew Gaston. Creating xbd: A dataset for assessing building damage from satellite imagery. In Pro- ceedings of the IEEE/CVF conference on computer ...

    2019

  6. [14]

    Continuous urban change detection from satellite image time series with temporal feature refinement and multi-task integration

    Sebastian Hafner, Heng Fang, Hossein Azizpour, and Yifang Ban. Continuous urban change detection from satellite image time series with temporal feature refinement and multi-task integration. arXiv preprint arXiv:2406.17458, 2024. 1, 2

    2024 arXiv

  7. [15]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll´ar, and Ross Girshick. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 16000– 16009, 2022. 11

    2022

  8. [16]

    Eurosat: A novel dataset and deep learning benchmark for land use and land cover classification

    Patrick Helber, Benjamin Bischke, Andreas Dengel, and Damian Borth. Eurosat: A novel dataset and deep learning benchmark for land use and land cover classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 12(7):2217–2226, 2019. 5

    2019

  9. [17]

    Supervised contrastive learning

    Prannay Khosla, Piotr Teterwak, Chen Wang, Aaron Sarna, Yonglong Tian, Phillip Isola, Aaron Maschinot, Ce Liu, and Dilip Krishnan. Supervised contrastive learning. Advances in neural information processing systems, 33:18661–18673,

  10. [18]

    Ai foundation models in remote sensing: A survey

    Siqi Lu, Junlin Guo, James R Zimmer-Dauphinee, Jordan M Nieusma, Xiao Wang, Parker VanValkenburgh, Steven A Wernke, and Yuankai Huo. Ai foundation models in remote sensing: A survey. arXiv preprint arXiv:2408.03464, 2024. 2, 4

    2024 arXiv

  11. [19]

    Change- aware sampling and contrastive learning for satellite images

    Utkarsh Mall, Bharath Hariharan, and Kavita Bala. Change- aware sampling and contrastive learning for satellite images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 5261–5270, 2023. 2, 3

    2023

  12. [20]

    Seasonal contrast: Un- supervised pre-training from uncurated remote sensing data

    Oscar Manas, Alexandre Lacoste, Xavier Gir ´o-i Nieto, David Vazquez, and Pau Rodriguez. Seasonal contrast: Un- supervised pre-training from uncurated remote sensing data. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 9414–9423, 2021. 2

    2021

  13. [21]

    Pangaea: A global and inclusive benchmark for geospatial foundation models

    Valerio Marsocci, Yuru Jia, Georges Le Bellier, David Kerekes, Liang Zeng, Sebastian Hafner, Sebastian Gerard, Eric Brune, Ritu Yadav, Ali Shibli, et al. Pangaea: A global and inclusive benchmark for geospatial foundation models. arXiv preprint arXiv:2412.04204, 2024. 2

    2024 arXiv

  14. [22]

    Pytorch: An im- perative style, high-performance deep learning library

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An im- perative style, high-performance deep learning library. Ad- vances in neural information processing systems ...

    2019

  15. [23]

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

    Colorado J Reed, Ritwik Gupta, Shufan Li, Sarah Brock- man, 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 Proceedings of th...

    2023

  16. [24]

    Ravæn: unsupervised change detection of extreme events using ml on-board satel- lites

    V ´ıt R ˘uˇziˇcka, Anna Vaughan, Daniele De Martini, James Fulton, Valentina Salvatelli, Chris Bridges, Gonzalo Mateo- Garcia, and Valentina Zantedeschi. Ravæn: unsupervised change detection of extreme events using ml on-board satel- lites. Scientific reports, 12(1):16939, 202...

    2022

  17. [25]

    Re- cent advances in autoencoder-based representation learning

    Michael Tschannen, Olivier Bachem, and Mario Lucic. Re- cent advances in autoencoder-based representation learning. arXiv preprint arXiv:1812.05069, 2018. 4

    2018 arXiv

  18. [26]

    Foundation models for remote sensing and earth observation: A survey

    Aoran Xiao, Weihao Xuan, Junjue Wang, Jiaxing Huang, Dacheng Tao, Shijian Lu, and Naoto Yokoya. Foundation models for remote sensing and earth observation: A survey. arXiv preprint arXiv:2410.16602, 2024. 2

    2024 arXiv

  19. [27]

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

    Zhitong Xiong, Yi Wang, Fahong Zhang, Adam J Stewart, Jo¨elle 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 e-prints, pages arXiv–...

    2024

  20. [28]

    Unsupervised flood detection on sar time series us- ing variational autoencoder.International Journal of Applied Earth Observation and Geoinformation , 126:103635, 2024

    Ritu Yadav, Andrea Nascetti, Hossein Azizpour, and Yifang Ban. Unsupervised flood detection on sar time series us- ing variational autoencoder.International Journal of Applied Earth Observation and Geoinformation , 126:103635, 2024. 1

    2024

  21. [29]

    Building damage assessment for rapid dis- aster response with a deep object-based semantic change de- tection framework: From natural disasters to man-made dis- asters

    Zhuo Zheng, Yanfei Zhong, Junjue Wang, Ailong Ma, and Liangpei Zhang. Building damage assessment for rapid dis- aster response with a deep object-based semantic change de- tection framework: From natural disasters to man-made dis- asters. Remote Sensing of Environment, 265:112...

    2021

  22. [30]

    Towards trans- ferable building damage assessment via unsupervised single- temporal change adaptation

    Zhuo Zheng, Yanfei Zhong, Liangpei Zhang, Marshall Burke, David B Lobell, and Stefano Ermon. Towards trans- ferable building damage assessment via unsupervised single- temporal change adaptation. Remote Sensing of Environ- ment, 315:114416, 2024. 1 10 Appendix A. Algorithm to ...

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.