Pith. sign in

REVIEW 3 major objections 6 minor 45 references

GeoFormer: A Multi-Polygon Segmentation Transformer

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read GeoFormer generates building polygons directly as autoregressive token sequences and reports AP 91.5 on the Aicrowd Mapping Challenge, about 12 points above the previous best.

desk verdict First autoregressive multi-polygon transformer for satellite buildings with a big claimed SOTA jump, but the evaluation protocol is underspecified enough that the 12-point gap is not yet verifiable. read the letter →

arxiv 2411.16616 v1 pith:EJJBGLY3 submitted 2024-11-25 cs.CV

Maxim Khomiakov , Michael Riis Andersen , Jes Frellsen This is my paper · ORCID
classification cs.CV
keywords GeoFormerbuildingfootprintextractionautoregressivetransformermulti-polygonsegmentationsatelliteimagerySWINv2AicrowdMappingChallengesequencegeneration
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

GeoFormer argues that building footprints in satellite imagery should be produced directly as vectors rather than as segmentation masks that later need polygonization. It presents an encoder-decoder transformer that flattens every building polygon in an image into one sequence of discretized x and y coordinates, with special tokens for the start of the sequence, the separation between buildings, and the end. The whole model is trained with a single categorical likelihood, so there are no loss-term weights to tune and no raster-to-vector post-processing. On the Aicrowd Mapping Challenge validation set the model reports an AP of 91.5, about 12 points above the previous best, and leads on boundary AP and complexity-aware IoU. The paper's claim, read sympathetically, is that scale-invariant building vectorization is a sequence-generation problem that transformers can solve end-to-end.

What carries the argument

The central object is a token sequence: all vertices of all buildings in an image are flattened as $(x_0,y_0,x_1,y_1,\dots)$ in a fixed order, separated by a separator token between buildings and bounded by start and stop tokens. The decoder computes the conditional likelihood $p_\theta(s_t|I,s_{<t})$ over discretized coordinates plus three special tokens, using causal self-attention over the sequence and cross-attention to a $36 \times 36$ feature map built from the four SWINv2 encoder stages. Two positional mechanisms do much of the work: RoPE encodes relative positions between coordinate tokens, and ALiBi gives a third of the attention heads a linear bias toward nearby tokens so the model can localize each object's vertices.

What would settle it

Re-run FFL, PolyWorld, and HiSup through the exact evaluation pipeline used for GeoFormer, including matching only polygons with more than 50% overlap and decoding with nucleus sampling p=0.95; if the AP margin over HiSup falls well below the reported 12 points, the central comparison is an artifact of protocol rather than the model. A second check is to set inference to greedy decoding and count how many sampled sequences per image are needed to reproduce the reported AP and AR.

Watch

Extended reading notes

Core claim

GeoFormer establishes that multi-building vectorization works as autoregressive polygon generation: the decoder predicts each vertex coordinate conditional on the satellite image and all previously predicted tokens. Modeling keypoints as spatially dependent tokens lets a single SWINv2-based encoder-decoder transformer output closed polygons directly, with no bounding-box pretraining, no separate corner detection, and no learned or heuristic polygonization stage. The reported results on the Aicrowd Mapping Challenge are AP 91.5, AR 97.8, bAP 97.1, IoU 98.1, and C-IoU 97.4, compared with HiSup's 79.4, 81.5, 66.5, 94.3, and 89.6; the only metric where GeoFormer trails is PoLiS distance, 0.913 versus 0.726. The paper further claims this is the first successful application of autoregressive transformer models to multi-polygon prediction in remote sensing.

Load-bearing premise

The performance comparison assumes the earlier methods' published scores were obtained under the same polygon-matching, filtering, and inference-sampling rules that the authors apply to GeoFormer, since the paper does not re-run those baselines with its own pipeline.

Editorial extensions

If this is right

  • A single negative-log-likelihood objective replaces the common combination of segmentation loss, polygonization loss, and hand-tuned weights.
  • Multi-building scenes are handled by a separator token, so the approach is not limited to single-object prediction like earlier Polygon-RNN variants.
  • The model's bAP and C-IoU advantages indicate that directly predicting vertices yields cleaner edges, which matters for map products that need right angles and precise corners.
  • The performance gain currently costs about 1.93 seconds per image with 97.4M parameters, roughly 64 times slower than HiSup, so speed is the main practical price.
  • Robustness results show the model is comparatively stable under rotation and missing pixels but loses more than HiSup under spatial downsampling, consistent with the $36 \times 36$ feature map being a resolution bottleneck.

Reading between the lines

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

  • The same sequence parameterization should extend to other vector map objects such as roads, parcels, or building clusters, and to wireframe-like problems where shared vertices must be revisited; the paper names wireframes as future work, but the transfer to other vector layers is not explored.
  • A testable consequence of the categorical tokenization is that switching to a discretized mixture-of-logistics likelihood, which the paper suggests for future work, could shorten sequences and reduce sampling cost while possibly changing the reported AP.
  • The paper leaves unquantified how much of the 12-point gap is architectural versus procedural; re-evaluating all baselines with identical 50%-overlap matching and identical inference sampling would separate those factors.
  • Because the reported numbers come from nucleus sampling with p=0.95, decoding greedily or varying the number of sampled sequences per image would reveal how much of the high AP depends on stochastic inference rather than the learned distribution.
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 / 6 minor

Summary. The paper proposes GeoFormer, an encoder-decoder autoregressive transformer that directly generates multiple building polygons from satellite imagery by representing polygons as flattened sequences of discretized x/y coordinate tokens with start, separator, and stop tokens, and by optimizing a single negative log-likelihood. The encoder is a SwinV2 feature-pyramid network, and the decoder combines causal attention, cross-attention, RoPE embeddings, and ALiBi biases. The authors report state-of-the-art results on the Aicrowd Mapping Challenge, with AP 91.5 versus HiSup's 79.4, and include ablations over 32 configurations plus robustness experiments under downsampling, dropout, and rotation perturbations.

Significance. If the reported results are reproducible, GeoFormer is a meaningful contribution: it is one of the first demonstrations that a single likelihood-based autoregressive transformer can predict multiple building polygons end-to-end, without per-loss weights or raster-to-vector post-processing. The paper ships code, reports a large set of ablations, and makes a falsifiable performance claim. The main weakness is that the central state-of-the-art claim currently rests on an underspecified evaluation protocol and on baseline numbers that were not produced with the same pipeline, so the 12-point AP improvement cannot yet be verified.

major comments (3)
  1. [Section 5, Table 1, Section 4.1] The headline state-of-the-art claim is not verifiable as written because AP and AR are rank-based metrics, but the paper never specifies how GeoFormer's stochastically sampled polygons are scored or ordered. Section 5 states only that inference uses nucleus sampling with p=0.95; it does not state how many sequences are sampled per image, whether each polygon is scored by sequence likelihood, average token log-probability, a special-token probability, or a constant score, or how duplicate or overlapping sampled polygons are removed. Please provide the full inference protocol (number of samples, scoring rule, deduplication, and rank ordering) and release the exact evaluation code used for the GeoFormer column of Table 1.
  2. [Table 1, Section 4.1] The comparison with baselines is not made under a common evaluation pipeline. The table footnote says baseline values were provided by the respective paper authors or by HiSup's authors, while GeoFormer's numbers were produced by the authors' own pipeline, which, for bAP, C-IoU, and PoLiS, first filters candidate polygons by requiring more than 50% overlap with the ground truth. Because the same filter and rasterization details may or may not have been used for the baseline numbers, the 12-percentage-point AP gap cannot be taken at face value. The authors should rerun all baselines with their own evaluation script and report the results, or, failing that, restrict the SOTA claim to the GeoFormer pipeline and clearly mark which published numbers were used.
  3. [Abstract, Section 2] The claim that this is the first successful application of auto-regressive transformer models for multi-polygon predictions in remote sensing is not substantiated. The related-work discussion dismisses earlier auto-regressive studies, including Polygonizer [15], as "limited to predicting a single object per scene," but no details or results for Polygonizer are given, so the reader cannot verify whether Polygonizer already addressed multi-polygon prediction or used a transformer decoder. Please state explicitly how Polygonizer and other auto-regressive methods differ in output space and architecture, and clarify whether they were evaluated on the same benchmark.
minor comments (6)
  1. [Section 4.1] The N-ratio is described only as comparing the cardinality of the predicted polygon to the ground truth; please provide the exact formula and explain how it is aggregated over a test set.
  2. [Table 2 and Table 5] The main-text ablation table consists of rows of checkmarks without a legend or row labels, so it is impossible to identify the ablated configuration from the printed table. Table 5 is dense and lacks clear row labeling; please add explicit row labels or configuration keys to both tables.
  3. [Equation (8) and reference [12]] Equation (8) uses a Sigmoid Linear Unit (SiLU) activation but cites reference [12], which is the GELU paper; please correct the citation or the activation name.
  4. [Section 3.1] The feature-map dimension list begins with H x W and the final representation is said to be 36 x 36 x C, but this is not derivable from a 224 x 224 input with patch size 4; please clarify the pyramid fusion and upsampling stages that produce the 36 x 36 feature map.
  5. [Appendix A] The optimizer is specified as beta = (0.9, 0.999), but the manuscript should state which value is beta_1 and which is beta_2, or use the standard notation beta = (beta_1, beta_2).
  6. [First page] There are spacing artifacts in the title and affiliation text ("T echnical", "Pol YGON", "SEGMENTA TION"); please clean the LaTeX source.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: GeoFormer's results are an empirical benchmark evaluation with no prediction that reduces to a fitted input or self-citation chain.

full rationale

The paper does not present a derivation in which a claimed prediction is equivalent to its inputs by construction. The method is an autoregressive transformer trained by maximum likelihood (Eq. 3), and the headline results in Table 1 are measured against the held-out Aicrowd validation set using external baselines. The baseline numbers are taken from the respective papers or from HiSup's authors, while GeoFormer's numbers come from the authors' own inference pipeline; this is a legitimate evaluation-consistency concern, but not circularity, because GeoFormer's parameters are trained on the training split and its reported metrics are not fitted to the test set. The self-citations [15, 16] appear only in the related-work discussion of similar autoregressive polygon and mesh generation approaches; they are not load-bearing for the main result, and the paper does not invoke any uniqueness theorem or prior-work ansatz to force its design choices. The ablations (Table 2 and Table 5) empirically test individual components and do not rename a known empirical pattern as a new contribution. Therefore no circular step is present, and the appropriate score is 0.

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

This is an empirical deep learning paper, so there are no invented physical entities. The main assumptions are about data correctness and the fairness of the comparison. The listed free parameters are evaluation and design choices that directly affect the reported metrics.

free parameters (3)
  • nucleus_sampling_p = 0.95
    Chosen for inference; affects the trade-off between precision and recall. The paper does not justify this value with a sweep.
  • feature_map_size = 36x36
    Design choice for the encoded spatial resolution; the authors speculate it causes poor downsampling robustness.
  • matching_overlap_threshold = 0.5 IoU
    Used to match predicted polygons to ground truth before computing bAP, C-IoU, and PoLiS. This threshold is set explicitly and differs from typical COCO evaluation which uses a range of IoU thresholds.
assumptions (3)
  • domain assumption The Aicrowd Mapping Challenge dataset annotations are accurate and consistently labeled.
    All training and evaluation rely on the correctness of the provided building polygon labels.
  • domain assumption The baseline results from prior papers were computed on the same test set (Aicrowd competition validation set) and with comparable evaluation metrics.
    The paper takes baseline numbers from the respective papers or from HiSup authors without re-evaluating them with the same code. If the test splits or metric implementations differ, the comparison is unfair.
  • ad hoc to paper The described hyperparameters (e.g., depths, heads, learning rate) are sufficient for the model to converge to the reported performance.
    No hyperparameter search is reported; the configuration appears to be a single hand-selected setting.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GeoFormer: A Multi-Polygon Segmentation Transformer." pith.science (2026). https://pith.science/paper/EJJBGLY3

@misc{pith2026241116616,
  author       = {Pith},
  title        = {Pith review of: GeoFormer: A Multi-Polygon Segmentation Transformer},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EJJBGLY3}},
  note         = {Machine review of arXiv:2411.16616}
}
read the original abstract

In remote sensing there exists a common need for learning scale invariant shapes of objects like buildings. Prior works relies on tweaking multiple loss functions to convert segmentation maps into the final scale invariant representation, necessitating arduous design and optimization. For this purpose we introduce the GeoFormer, a novel architecture which presents a remedy to the said challenges, learning to generate multipolygons end-to-end. By modeling keypoints as spatially dependent tokens in an auto-regressive manner, the GeoFormer outperforms existing works in delineating building objects from satellite imagery. We evaluate the robustness of the GeoFormer against former methods through a variety of parameter ablations and highlight the advantages of optimizing a single likelihood function. Our study presents the first successful application of auto-regressive transformer models for multi-polygon predictions in remote sensing, suggesting a promising methodological alternative for building vectorization.

Figures

Figures reproduced from arXiv: 2411.16616 by the authors.

Figure 1
Figure 1. Illustration of the GeoFormer model architecture. On the left hand side an image [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Qualitative results of model predictions on test set images together with ground [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Visual examples of perturbations performed to input images in the robustness [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Performance relative to perturbations performed on the Aicrowd small dataset. We [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]
Figure 5
Figure 5. Figure 5: Visualisation of the attention maps on top of the input image and predicted polygons [PITH_FULL_IMAGE:figures/full_fig_p017_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 26 canonical work pages

  1. [15]

    Polygonizer: An auto-regressive building delineator

    Maxim Khomiakov, Michael Riis Andersen, and Jes Frellsen. Polygonizer: An auto- regressive building delineator. arXiv preprint arXiv:2304.04048, 2023

    work page Pith review arXiv 2023

  2. [1]

    Efficient Interactive Annotation of Segmentation Datasets with Polygon-RNN++

    David Acuna, Huan Ling, Amlan Kar, and Sanja Fidler. Efficient Interactive Annotation of Segmentation Datasets with Polygon-RNN++. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pages 859–868, 2018. ISSN 10636919. doi: 10.1109/CVPR.2018.00096

    arXiv 2018

  3. [2]

    Alidoost, H

    F. Alidoost, H. Arefi, and F. Tombari. Building outline extraction from aerial images using convolutional neural networks. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, 42(4/W18):57–61,

    work page

  4. [3]

    A metric for polygon comparison and building extraction evaluation

    Janja Avbelj, Rupert Müller, and Richard Bamler. A metric for polygon comparison and building extraction evaluation. IEEE Geoscience and Remote Sensing Letters, 12(1): 170–174, 2015. doi: 10.1109/LGRS.2014.2330695

    arXiv 2015

  5. [4]

    Layer normalization

    Jimmy Lei Ba, Jamie Ryan Kiros, and Geoffrey E Hinton. Layer normalization. arXiv preprint arXiv:1607.06450, 2016

    arXiv 2016

  6. [5]

    Building Footprint Extraction From VHR Remote Sensing Images Combined With Normalized DSMs Using Fused Fully Convolutional Networks

    Ksenia Bittner, Fathalrahman Adam, Shiyong Cui, Marco Körner, and Peter Reinartz. Building Footprint Extraction From VHR Remote Sensing Images Combined With Normalized DSMs Using Fused Fully Convolutional Networks. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11(8):2615–2629,

    work page

  7. [6]

    Heat: Holistic edge attention transformer for structured reconstruction

    Jiacheng Chen, Yiming Qian, and Yasutaka Furukawa. Heat: Holistic edge attention transformer for structured reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3866–3875, 2022

    work page 2022

  8. [7]

    A unified sequence interface for vision tasks

    Ting Chen, Saurabh Saxena, Lala Li, Tsung-Yi Lin, David J Fleet, and Geoffrey E Hinton. A unified sequence interface for vision tasks. Advances in Neural Information Processing Systems, 35:31333–31346, 2022

    2022

Show all 45 references
  1. [8]

    Boundary iou: Improving object-centric image segmentation evaluation

    Bowen Cheng, Ross Girshick, Piotr Dollár, Alexander C Berg, and Alexander Kirillov. Boundary iou: Improving object-centric image segmentation evaluation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 15334–15342, 2021

    2021

  2. [9]

    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. arXiv preprint arXiv...

    2010 arXiv

  3. [10]

    Polygonal Building Segmentation by Frame Field Learning

    Nicolas Girard, Dmitriy Smirnov, Justin Solomon, and Yuliya Tarabalka. Polygonal Building Segmentation by Frame Field Learning. pages 1–30, 2020. URL http: //arxiv.org/abs/2004.14875

    2020 arXiv

  4. [11]

    Polygonal building extraction by frame field learning

    Nicolas Girard, Dmitriy Smirnov, Justin Solomon, and Yuliya Tarabalka. Polygonal building extraction by frame field learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 5891–5900, 2021. GEOFORMER: A MULTI-POL YGON SEGMENTA TION ...

    2021

  5. [12]

    Gaussian error linear units (gelus)

    Dan Hendrycks and Kevin Gimpel. Gaussian error linear units (gelus). arXiv preprint arXiv:1606.08415, 2016

    2016 arXiv

  6. [13]

    The curious case of neural text degeneration

    Ari Holtzman, Jan Buys, Li Du, Maxwell Forbes, and Yejin Choi. The curious case of neural text degeneration. arXiv preprint arXiv:1904.09751, 2019

    1904 arXiv

  7. [14]

    Polybuilding: Polygon transformer for end-to-end building extraction

    Yuan Hu, Zhibin Wang, Zhou Huang, and Yu Liu. Polybuilding: Polygon transformer for end-to-end building extraction. arXiv preprint arXiv:2211.01589, 2022

    2022 arXiv

  8. [16]

    Learning to generate 3d representations of building roofs using single-view aerial imagery

    Maxim Khomiakov, Alejandro Valverde Mahou, Alba Reinders Sánchez, Jes Frellsen, and Michael Riis Andersen. Learning to generate 3d representations of building roofs using single-view aerial imagery. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Sig...

    2023

  9. [17]

    Joint semantic- geometric learning for polygonal building segmentation

    Weijia Li, Wenqian Zhao, Huaping Zhong, Conghui He, and Dahua Lin. Joint semantic- geometric learning for polygonal building segmentation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 1958–1965, 2021

    1958

  10. [18]

    Joint semantic–geometric learning for polygonal building segmentation from high-resolution remote sensing images

    Weijia Li, Wenqian Zhao, Jinhua Yu, Juepeng Zheng, Conghui He, Haohuan Fu, and Dahua Lin. Joint semantic–geometric learning for polygonal building segmentation from high-resolution remote sensing images. ISPRS Journal of Photogrammetry and Remote Sensing, 201:26–37, 2023

    2023

  11. [19]

    Topological map extraction from overhead images

    Zuoyue Li, Jan DIrk Wegner, and Aurelien Lucchi. Topological map extraction from overhead images. Proceedings of the IEEE International Conference on Computer Vision, 2019-Octob:1715–1724, 2019. ISSN 15505499. doi: 10.1109/ICCV .2019.00180

    2019

  12. [20]

    Microsoft coco: Common objects in context

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ra- manan, Piotr Dollár, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzer- land, September 6-12, 2014, Proceed...

    2014

  13. [21]

    Feature pyramid networks for object detection

    Tsung-Yi Lin, Piotr Dollár, Ross Girshick, Kaiming He, Bharath Hariharan, and Serge Belongie. Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 2117–2125, 2017

    2017

  14. [22]

    Polyformer: Referring image segmentation as sequential polygon generation

    Jiang Liu, Hui Ding, Zhaowei Cai, Yuting Zhang, Ravi Kumar Satzoda, Vijay Ma- hadevan, and R Manmatha. Polyformer: Referring image segmentation as sequential polygon generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18653–1...

    2023

  15. [23]

    Swin transformer v2: Scaling up capacity and resolution

    Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, et al. Swin transformer v2: Scaling up capacity and resolution. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 12009–12019, 2022

    2022

  16. [24]

    Learning deep structured active contours end-to-end

    Diego Marcos, Devis Tuia, Benjamin Kellenberger, Lisa Zhang, Min Bai, Renjie Liao, and Raquel Urtasun. Learning deep structured active contours end-to-end. In Pro- ceedings of the IEEE conference on computer vision and pattern recognition , pages 8877–8885, 2018. 20 GEOFORMER:...

    2018

  17. [25]

    Deep learning for understanding satellite imagery: An experimental survey

    Sharada Prasanna Mohanty, Jakub Czakon, Kamil A Kaczmarek, Andrzej Pyskir, Piotr Tarasiewicz, Saket Kunwar, Janick Rohrbach, Dave Luo, Manjunath Prasad, Sascha Fleer, et al. Deep learning for understanding satellite imagery: An experimental survey. Frontiers in Artificial Inte...

    2020

  18. [26]

    Charlie Nash, Yaroslav Ganin, S. M. Ali Eslami, and Peter W. Battaglia. Polygen: An autoregressive generative model of 3d meshes, 2020

    2020

  19. [27]

    Train short, test long: Attention with linear biases enables input length extrapolation

    Ofir Press, Noah A Smith, and Mike Lewis. Train short, test long: Attention with linear biases enables input length extrapolation. arXiv preprint arXiv:2108.12409, 2021

    2021 arXiv

  20. [28]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer- Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part...

    2015

  21. [29]

    Pixelcnn++: Improv- ing the pixelcnn with discretized logistic mixture likelihood and other modifications

    Tim Salimans, Andrej Karpathy, Xi Chen, and Diederik P Kingma. Pixelcnn++: Improv- ing the pixelcnn with discretized logistic mixture likelihood and other modifications. arXiv preprint arXiv:1701.05517, 2017

    2017 arXiv

  22. [30]

    Roformer: Enhanced transformer with rotary position embedding

    Jianlin Su, Yu Lu, Shengfeng Pan, Ahmed Murtadha, Bo Wen, and Yunfeng Liu. Roformer: Enhanced transformer with rotary position embedding. arXiv preprint arXiv:2104.09864, 2021

    2021 arXiv

  23. [31]

    Annotating object instances with a polygon-RNN Castrejón et al

    Raquel Urtasun and Sanja Fidler. Annotating object instances with a polygon-RNN Castrejón et al. Computer Vision and Pattern Recognition, 2017. URL http://www. cs.toronto.edu/

    2017

  24. [32]

    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. Advances in neural information processing systems, 30, 2017

    2017

  25. [33]

    Show and tell: A neural image caption generator

    Oriol Vinyals, Alexander Toshev, Samy Bengio, and Dumitru Erhan. Show and tell: A neural image caption generator. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3156–3164, 2015

    2015

  26. [34]

    Hisup: Accurate polygonal mapping of buildings in satellite imagery with hierarchical supervision

    Bowen Xu, Jiakun Xu, Nan Xue, and Gui-Song Xia. Hisup: Accurate polygonal mapping of buildings in satellite imagery with hierarchical supervision. ISPRS Journal of Photogrammetry and Remote Sensing, 198:284–296, 2023

    2023

  27. [35]

    Show, attend and tell: Neural image caption generation with visual attention

    Kelvin Xu, Jimmy Ba, Ryan Kiros, Kyunghyun Cho, Aaron Courville, Ruslan Salakhudi- nov, Rich Zemel, and Yoshua Bengio. Show, attend and tell: Neural image caption generation with visual attention. In International conference on machine learning, pages 2048–2057. PMLR, 2015

    2015

  28. [36]

    Ferret: Refer and ground anything anywhere at any granularity, 2023

    Haoxuan You, Haotian Zhang, Zhe Gan, Xianzhi Du, Bowen Zhang, Zirui Wang, Liangliang Cao, Shih-Fu Chang, and Yinfei Yang. Ferret: Refer and ground anything anywhere at any granularity, 2023

    2023

  29. [37]

    Hit: Building mapping with hierarchical transformers

    Mingming Zhang, Qingjie Liu, and Yunhong Wang. Hit: Building mapping with hierarchical transformers. arXiv preprint arXiv:2309.09643, 2023. GEOFORMER: A MULTI-POL YGON SEGMENTA TION TRANSFORMER 21

    2023 arXiv

  30. [38]

    Building outline delineation: From aerial images to polygons with an improved end-to-end learning framework

    Wufan Zhao, Claudio Persello, and Alfred Stein. Building outline delineation: From aerial images to polygons with an improved end-to-end learning framework. ISPRS journal of photogrammetry and remote sensing, 175:119–131, 2021

    2021

  31. [39]

    Random erasing data augmentation

    Zhun Zhong, Liang Zheng, Guoliang Kang, Shaozi Li, and Yi Yang. Random erasing data augmentation. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 13001–13008, 2020

    2020

  32. [40]

    De- formable detr: Deformable transformers for end-to-end object detection

    Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, and Jifeng Dai. De- formable detr: Deformable transformers for end-to-end object detection. arXiv preprint arXiv:2010.04159, 2020

    2010 arXiv

  33. [41]

    Machine-learned Regular- ization and Polygonization of Building Segmentation Masks

    Stefano Zorzi, Ksenia Bittner, and Friedrich Fraundorfer. Machine-learned Regular- ization and Polygonization of Building Segmentation Masks. 2020. URL http: //arxiv.org/abs/2007.12587

    2020 arXiv

  34. [42]

    Poly- world: Polygonal building extraction with graph neural networks in satellite images

    Stefano Zorzi, Shabab Bazrafkan, Stefan Habenschuss, and Friedrich Fraundorfer. Poly- world: Polygonal building extraction with graph neural networks in satellite images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1848–1857, 2022

    2022

  35. [43]

    Generalized decoding for pixel, image, and language

    Xueyan Zou, Zi-Yi Dou, Jianwei Yang, Zhe Gan, Linjie Li, Chunyuan Li, Xiyang Dai, Harkirat Behl, Jianfeng Wang, Lu Yuan, et al. Generalized decoding for pixel, image, and language. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1511...

    2023

  36. [2018]

    doi: 10.1109/JSTARS.2018.2849363

    ISSN 21511535. doi: 10.1109/JSTARS.2018.2849363

    2018

  37. [2019]

    doi: 10.5194/isprs-archives-XLII-4-W18-57-2019

    ISSN 16821750. doi: 10.5194/isprs-archives-XLII-4-W18-57-2019

    2019 doi

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.