Pith. sign in

REVIEW 4 major objections 4 minor 56 references

CoLR-Det claims super-resolution can improve small-object detection as a training-only latent regularizer, not an image enhancer.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

CoLR-Det is announced in the abstract, but the supplied full text is actually the authors' SDCoNet paper, so the claimed result is not present in the artifact.

T0 review reviewed 2026-08-03 challenge →

load-bearing objection The submission is two different papers: the abstract describes CoLR-Det, the full text is a complete SDCoNet paper, so the claimed contribution is unsupported as submitted. the 4 major comments →

arxiv 2601.12507 v2 pith:XOLGCXSN submitted 2026-01-18 cs.CV cs.LG

CoLR-Det: Collaborative Latent Restoration for Small Object Detection in Low-Resolution Remote Sensing Images

Ruo Qi , Linhui Dai , Yusong Qin , Chaolei Yang , Yanshan Li This is my paper
classification cs.CV cs.LG
keywords remote sensing object detectionsmall object detectionsuper-resolutionlatent regularizationmulti-task learninglow-resolution imagerytoken routingDETR
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The reading

The paper argues that the standard "restore first, then detect" pipeline for low-resolution remote sensing is misaligned: super-resolution amplifies texture and edges, while detection needs object semantics, so restored images can be visually sharper yet semantically misleading. It proposes CoLR-Det, which instead runs a super-resolution branch only during training, imposing reconstruction constraints on shared multiscale features and then discarding the branch at inference, so the deployment path is purely detection-driven. The claimed outcome is higher small-object detection accuracy on three resolution-degraded benchmark datasets without paying super-resolution compute at inference. A caveat is that the supplied full text actually describes a differently named architecture (SDCoNet) with the same broad strategy; the CoLR-Det-specific experiments and ablations asserted in the abstract are not present in the provided manuscript.

Core claim

On the paper's own terms, the discovery is that super-resolution supervision can be redefined as detection-oriented latent regularization: a training-only restoration decoder attached to a shared encoder forces the shared features to retain small-object details, and can then be removed without harming detection at inference. The accompanying mechanisms—saliency-guided token routing, which concentrates attention on high-saliency tokens and keeps bypassed tokens through a learnable background embedding, and a two-stage schedule that stabilizes detection semantics before adding conservative SR updates—are presented as the means to make this regularization work. The abstract reports state-of-the

What carries the argument

The central mechanism is the training-only restoration branch: a super-resolution decoder that shares a Swin-Transformer encoder with the detection head, contributes an auxiliary reconstruction loss during training, and is removed at inference. Two supporting mechanisms carry the argument: saliency-guided token routing (tokens are ranked by a predicted saliency score, attention-based refinement is applied only to high-saliency tokens, and low-saliency tokens are preserved with a learnable background embedding), and detection-prioritized two-stage optimization (detection is trained alone with the SR branch frozen, then the SR branch is unfrozen at a smaller learning rate). Together they conve

Load-bearing premise

The load-bearing premise is that training-only auxiliary reconstruction constraints on shared multiscale features improve detection accuracy once the super-resolution decoder is discarded at inference—a hypothesis the abstract asserts but the supplied full text (which describes a different architecture, SDCoNet) does not directly test.

What would settle it

Train the same detection head and shared encoder with the SR branch present versus absent (or with the SR loss weight set to zero) on the same low-resolution splits while keeping all other components fixed; if small-object AP does not clearly drop when the SR branch is removed, the latent-regularization claim is falsified. The paper's own ablations would constitute this experiment if they were run for CoLR-Det.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • If the latent-regularization claim holds, SR-assisted detection pipelines can remove the super-resolution network at inference entirely, eliminating its compute and latency while keeping the accuracy gain.
  • The framing implies reconstruction losses no longer need to be optimized for pixel fidelity (e.g., PSNR); only their effect on shared detection features matters, opening the door to cheaper or weaker restoration decoders.
  • The reported token-routing results suggest that in low-resolution remote sensing scenes, a large fraction of background tokens can be skipped without losing accuracy, which is a template for efficient Transformer-based detectors.
  • The two-stage training schedule indicates that freezing the auxiliary branch until detection semantics stabilize avoids gradient conflicts, a recipe that could transfer to other multi-task detection settings.

Where Pith is reading between the lines

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

  • If this works, the same 'auxiliary pixel task as latent regularizer' trick may generalize to other pixel-level losses—denoising, deblurring, inpainting—so detection on degraded imagery could be improved without any restoration module at test time.
  • A natural extension is to test whether the regularization effect is tied to the specific reconstruction objective or just to the presence of an extra training-time task; replacing the SR loss with a random feature-matching loss would separate those.
  • The mismatch between the CoLR-Det abstract and the SDCoNet full text means the manuscript as submitted does not actually contain the described architecture; the abstract's performance numbers should be treated as unverified until the matching full paper and code are available.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The submission as received is internally inconsistent. The title, author list, and abstract describe "CoLR-Det: Collaborative Latent Restoration for Small Object Detection in Low-Resolution Remote Sensing Images," whose central claim is that super-resolution supervision is used as a training-only latent regularizer on shared multiscale representations, with the SR decoder discarded at inference and no explicit image-level enhancement. However, the full text supplied is a complete manuscript for a different method, "SDCoNet: Saliency-Driven Multi-Task Collaborative Network for Remote Sensing Object Detection." Sections I-V, all equations (Eqs. 1-11), Algorithm 1, and all experimental tables describe SDCoNet and never mention CoLR-Det. The reported results (e.g., Table II: AP 0.836 on NWPU VHR-10-Split; Table III: AP 0.419 on DOTAv1.5-Split; Table IV: AP 0.701 on HRSSD-Split) are SDCoNet results under explicit image-level SR supervision, not CoLR-Det results. Therefore the paper's stated contribution—latent restoration as an implicit regularizer—is not defined, derived, or evaluated in the submitted artifact.

Significance. If the CoLR-Det claim were validated—that a training-only auxiliary reconstruction constraint on shared features, with the SR decoder removed at inference, improves small-object detection without explicit image-level enhancement—it would be a meaningful contribution to low-resolution remote sensing detection. The existing SDCoNet body does provide some useful empirical material: Table V shows that adding an SR branch to Cascade R-CNN and DINO improves AP and especially AP_s on NWPU VHR-10-Split and DOTAv1.5-Split, with the SR head not used at inference, and Tables II-IV compare against several recent detectors under a controlled low-resolution protocol. These results give partial evidence for the broader idea that auxiliary SR supervision can help detection. However, none of this supports the specific latent-restoration mechanism claimed in the abstract. The artifact contains no derivation, algorithm, ablation, or experiment for CoLR-Det; the claimed mechanism is absent, and the one existing ablation (Table V) is for explicit pixel-level SR reconstruction (Eqs. 4 and 11), which is in tension with the abstract's claim of avoiding explicit image-level enhancement.

major comments (4)
  1. [Abstract vs. full text; Sections III-V] The manuscript body is SDCoNet, not CoLR-Det. The term "CoLR-Det" does not appear in the method section (Sections III.A-III.E), Algorithm 1, experimental section (Section IV), or conclusion (Section V). The equations describe SDCoNet's shared-encoder, U-Net SR decoder, saliency query filtering, and two-stage gradient routing. Consequently, none of the central CoLR-Det components—collaborative latent restoration, training-only restoration branch, or detection-oriented latent regularization—are defined or tested. The abstract's claim that CoLR-Det "outperforms state-of-the-art methods" is unsupported by any CoLR-Det experiment. This is not a presentation issue; the paper as submitted does not contain its claimed method.
  2. [Eq. (4), Eq. (11), Section III.C, Section III.E] The claimed latent-restoration mechanism contradicts the actual equations in the body. The abstract states that CoLR-Det treats SR supervision as "detection-oriented latent regularization rather than explicit image-level enhancement." But SDCoNet's SR branch explicitly reconstructs a high-resolution image: Eq. (4) defines I_SR = Conv(PixelShuffle(σ(Conv(F_dec)))), and Eq. (11) includes the pixel-level reconstruction loss β_sr L_sr in the stage-two objective. No equation or loss term applies reconstruction supervision to hidden shared representations as a latent regularizer. Thus the core mechanism advertised in the abstract is absent from the supplied method.
  3. [Table V, Section IV.C] Table V is the only ablation involving removal of the SR branch, but it evaluates SDCoNet, not CoLR-Det. It shows AP improvements when the explicit SR branch is enabled during training (e.g., DINO on NWPU VHR-10-Split: AP 0.766→0.816; AP_s 0.487→0.552), with the SR head discarded at inference. This is evidence for the weaker, already known hypothesis that auxiliary SR training can help a detector. It does not test the CoLR-Det-specific claim that a latent-restoration branch without explicit image reconstruction outperforms explicit SR training, nor does it isolate the effect of "latent regularization" versus pixel-level loss. No table or figure in the manuscript instantiates CoLR-Det's proposed training-only restoration branch.
  4. [General assessment of soundness] Because the method and experiments belong to a different named framework, the internal consistency of the paper is broken at the level of its central claim. The reported SDCoNet results, even if technically valid, cannot be used to evaluate CoLR-Det. The manuscript would require a complete rewrite of the method, including a formal definition of the latent-restoration loss, an inference-time architecture, and experiments explicitly comparing CoLR-Det against SDCoNet and other baselines. This goes beyond what can be addressed through minor revisions.
minor comments (4)
  1. [Figure 4] The category distribution figures contain raw Unicode substitution strings such as "/uni00000014/uni00000011/..." instead of legible text labels, making Fig. 4 unreadable.
  2. [Table IV, HRSSD-Split row] In the SDCoNet row, the AP_l column reads "0.6190.680" (likely two numbers concatenated without a separator). This should be corrected.
  3. [Section IV.A, Equation (8)] The notation in Eq. (8), q_i = A(q_i + pos_i, q), is confusing: the attention operation uses q_i both as a query to be updated and as part of the argument list, while q is not prominently defined. Please clarify the query/key/value roles.
  4. [References] The reference list contains duplicates and inconsistencies: [55] is the same paper as [41]; [10] is cited as "SR4IR" but the title at the reference entry is "Beyond image super-resolution for image recognition with task-driven perceptual loss," which should be checked for accuracy.

Circularity Check

0 steps flagged

No circularity: the claimed latent-regularization effect is tested by an ablation, and no derivation reduces to its inputs.

full rationale

No circular derivation chain is present. No parameter is fitted to a subset and then reported as a prediction; no load-bearing self-citation; no imported uniqueness theorem; no ansatz smuggled via citation; no renaming of a known result. The main integrity concern is a front-matter/body mismatch: the abstract describes CoLR-Det and 'latent regularization rather than explicit image-level enhancement,' but the body presents SDCoNet, whose Eq. (4) reconstructs I_SR and Eq. (11) trains it with pixel loss L_sr. That is a self-consistency and support failure, not circularity. The central claim, that a training-only SR branch improves detection after the SR head is discarded, is directly tested in Table V (e.g., DINO AP 0.766 to 0.816 on NWPU VHR-10-Split) and is consistent with the statement 'since the SR head is discarded during inference.' Experiments are judged against external baselines, not against quantities defined from the method's own outputs. No equation reduces to an input by construction. Accordingly, the circularity score is 0; the mismatch should be treated as a correctness/reproducibility issue rather than circularity.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The only CoLR-Det premise that can be extracted is the latent-regularization hypothesis from the abstract; all other parameters and entities are un-auditable because the submitted full text is SDCoNet, not CoLR-Det.

free parameters (1)
  • CoLR-Det training hyperparameters (loss weights, SR learning-rate ratio, token retention ratios)
    Named in the abstract (smaller SR learning rate, saliency-guided token routing) but no numeric values are given in the artifact, and the body is SDCoNet so they cannot be recovered.
axioms (1)
  • domain assumption Training-only SR reconstruction loss on shared representations improves detection accuracy even though the SR decoder is discarded at inference (latent-regularization hypothesis).
    Central premise of the CoLR-Det abstract; no derivation or ablation is present in the supplied artifact.

reviewed 2026-08-03 · how reviews work

0 comments
Cite this review

Pith. "Pith review of CoLR-Det: Collaborative Latent Restoration for Small Object Detection in Low-Resolution Remote Sensing Images." pith.science (2026). https://pith.science/paper/XOLGCXSN

@misc{pith2026260112507,
  author       = {Pith},
  title        = {Pith review of: CoLR-Det: Collaborative Latent Restoration for Small Object Detection in Low-Resolution Remote Sensing Images},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XOLGCXSN}},
  note         = {Machine review of arXiv:2601.12507}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Low-resolution remote sensing small object detection is limited by both missing visual details and the ambiguity of how details serve detection. Existing super-resolution-assisted detectors generally follow a restoration-first paradigm to explicitly enhance inputs before detection, which implicitly assumes visual fidelity benefits recognition. Yet super-resolution favors dense texture and edge recovery, while object detection relies on sparse instance-level semantics, making restoration amplify visually plausible but semantically irrelevant background textures. To tackle this issue, we propose CoLR-Det, a Collaborative Latent-Restoration-Assisted Small Object Detection framework that treats super-resolution supervision as detection-oriented latent regularization rather than explicit image-level enhancement. Instead of reconstructing high-resolution images for inference, CoLR-Det uses a training-only restoration branch to impose auxiliary reconstruction constraints on shared multiscale representations, and the inference pathway remains purely detection-driven. We further design a saliency-guided object-preserving token routing mechanism, which prioritizes high-saliency tokens for attention-based refinement while retaining information of bypassed tokens. Besides, a detection-prioritized two-stage optimization strategy is developed: it first builds stable object-level semantics before introducing restoration supervision, and assigns a smaller learning rate to the SR decoder to keep its updates conservative and reduce perturbations in collaborative training. With this design, CoLR-Det transforms restoration from an explicit visual enhancement operator into an implicit semantic regularizer. Experiments on resolution-degraded NWPU VHR-10-Split, DOTAv1.5-Split and HRSSD-Split show that CoLR-Det outperforms state-of-the-art methods, with code available at https://github.com/qiruo-ya/CoLR-Det.

Figures

Figures reproduced from arXiv: 2601.12507 by Chaolei Yang, Linhui Dai, Ruo Qi, Yanshan Li, Yusong Qin.

Figure 1
Figure 1. Figure 1: Visualization of activation maps for different tasks [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The framework pipeline of SDCoNet. Our SDCoNet comprises two core branches: super-resolution and object detection, consisting of a shared encoder, a super-resolution decoder, a saliency-driven query filtering module, and an object detection encoder-decoder; in the first phase, the SR branch is frozen to update the object detection network, and in the second phase, dual-branch collaborative training is init… view at source ↗
Figure 3
Figure 3. Figure 3: Example images from the DOTAv1.5-Split [22], NWPU VHR-10-Split [21], and HRSSD-Split [23] datasets. The left side shows high-resolution images from these datasets, while the right side presents the corresponding low-resolution images generated by bicubic downsampling. 17.6% 16.0% 15.6% 13.9% 10.7% 8.1% 5.8% 4.4% 4.4% 3.4% airplane 17.6% tennis-court 16.0% storage-tank 15.6% vehicle 13.9% baseball-diamond 1… view at source ↗
Figure 4
Figure 4. Figure 4: Category distribution of the benchmark datasets. The pie charts show the class proportions of the DOTAv1.5-Split [22], NWPU VHR-10-Split [21], and HRSSD-Split [23] datasets, which contain a large number of small-object categories such as storage-tank, small-vehicle, bridge, and ship. only receives these LR images (512×512 or 256×256), which forces the model to recover fine details and detect objects from s… view at source ↗
Figure 5
Figure 5. Figure 5: Per-class detection performance on the NWPU VHR-10- Split dataset. The chart compares SDCoNet with representative CNN-based (Faster R-CNN, YOLOX-s, FFCA-YOLO, LEGNet-T) and Transformer-based (DINO, DN-DETR) detectors, showing consistent gains for small-object categories such as storage-tank and vehicle. CNN, RetinaNet, Cascade R-CNN, YOLOX-s, Sparse R-CNN, Deformable DETR, and DINO. On DOTAv1.5-Split, SDCo… view at source ↗
Figure 6
Figure 6. Figure 6: Qualitative detection results of the proposed SDCoNet compared with representative state-of-the-art detectors. From left to right, the columns correspond to Faster R-CNN, RetinaNet, Cascade R-CNN, YOLOX-s, Sparse R-CNN, Deformable DETR, DINO, and SDCoNet. From top to bottom, the rows show visualization results on the NWPU VHR-10-Split [21], DOTAv1.5-Split [22], and HRSSD-Split [23] datasets, respectively. … view at source ↗
Figure 7
Figure 7. Figure 7: Visualization of intermediate feature maps for the baseline DINO and the SR-augmented DINO-SR Branch on the NWPU VHR-10-Split [21] dataset. From left to right, the columns correspond to backbone stages 1–4; within each stage, the first rows show the activation maps of DINO, while the second rows show those of DINO-SR Branch. TABLE IX EFFICIENCY AND ACCURACY TRADE-OFFS UNDER DIFFERENT QUERY TOKEN FILTERING … view at source ↗

discussion (0)

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

Reference graph

Works this paper leans on

56 extracted references · 3 linked inside Pith

  1. [1]

    Small object detection: A comprehensive survey on challenges, techniques and real-world applications,

    M. Nikouei, B. Baroutian, S. Nabavi, F. Taraghi, A. Aghaei, A. Sajedi, and M. E. Moghaddam, “Small object detection: A comprehensive survey on challenges, techniques and real-world applications,”Intelligent Systems with Applications, vol. 27, p. 200561, 2025

    2025

  2. [2]

    Remote sensing object detection in the deep learning era—a review,

    S. Gui, S. Song, R. Qin, and Y . Tang, “Remote sensing object detection in the deep learning era—a review,”Remote Sensing, vol. 16, p. 327, 2024

    2024

  3. [3]

    A review of deep-learning-based super-resolution: From methods to applications,

    H. Su, Y . Li, Y . Xu, X. Fu, and S. Liu, “A review of deep-learning-based super-resolution: From methods to applications,”Pattern Recognition, vol. 157, p. 110935, 2025

    2025

  4. [4]

    Image super- resolution: A comprehensive review, recent trends, challenges and ap- plications,

    D. C. Lepcha, B. Goyal, A. Dogra, and V . Goyal, “Image super- resolution: A comprehensive review, recent trends, challenges and ap- plications,”Information Fusion, vol. 91, pp. 230–260, 2023

    2023

  5. [5]

    Lightweight remote sensing super-resolution with multi-scale graph attention network,

    Y . Wang, Z. Shao, T. Lu, X. Huang, J. Wang, Z. Zhang, and X. Zuo, “Lightweight remote sensing super-resolution with multi-scale graph attention network,”Pattern Recognition, vol. 160, p. 111178, 2025

    2025

  6. [6]

    A novel hybrid deep learning approach for super-resolution and objects detection in remote sensing,

    N. Alsaedi and B. Shah, “A novel hybrid deep learning approach for super-resolution and objects detection in remote sensing,”Scientific Reports, vol. 15, p. 1476, 2025

    2025

  7. [7]

    Task-driven super res- olution: Object detection in low-resolution images,

    M. Haris, G. Shakhnarovich, and N. Ukita, “Task-driven super res- olution: Object detection in low-resolution images,” inInternational Conference on Neural Information Processing. Springer, 2021, pp. 387–395

    2021

  8. [8]

    Distilling knowledge from super-resolution for efficient remote sensing salient object detection,

    Y . Liu, Z. Xiong, Y . Yuan, and Q. Wang, “Distilling knowledge from super-resolution for efficient remote sensing salient object detection,” IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1– 16, 2023

    2023

  9. [9]

    Small- object detection in remote sensing images with end-to-end edge- enhanced gan and object detector network,

    J. Rabbi, N. Ray, M. Schubert, S. Chowdhury, and D. Chao, “Small- object detection in remote sensing images with end-to-end edge- enhanced gan and object detector network,”Remote Sensing, vol. 12, no. 9, p. 1432, 2020

    2020

  10. [10]

    Beyond image super-resolution for image recognition with task-driven perceptual loss,

    J. Kim, J. Oh, and K. M. Lee, “Beyond image super-resolution for image recognition with task-driven perceptual loss,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 2651–2661

    2024

  11. [11]

    Dab-detr: Dynamic anchor boxes are better queries for detr. arxiv 2022,

    S. Liu, F. Li, H. Zhang, X. Yang, X. Qi, H. Su, J. Zhu, and L. Zhang, “Dab-detr: Dynamic anchor boxes are better queries for detr. arxiv 2022,”arXiv preprint arXiv:2201.12329

    Pith/arXiv arXiv 2022

  12. [12]

    Dn-detr: Accelerate detr training by introducing query denoising,

    F. Li, H. Zhang, S. Liu, J. Guo, L. M. Ni, and L. Zhang, “Dn-detr: Accelerate detr training by introducing query denoising,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 13 619–13 627. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 13

    2022

  13. [13]

    Deformable detr: Deformable transformers for end-to-end object detection,

    X. Zhu, W. Su, L. Lu, B. Li, X. Wang, and J. Dai, “Deformable detr: Deformable transformers for end-to-end object detection,”arXiv preprint arXiv:2010.04159, 2020

    Pith/arXiv arXiv 2010

  14. [14]

    Faster r-cnn: Towards real-time object detection with region proposal networks,

    S. Ren, K. He, R. Girshick, and J. Sun, “Faster r-cnn: Towards real-time object detection with region proposal networks,”IEEE transactions on pattern analysis and machine intelligence, vol. 39, no. 6, pp. 1137–1149, 2017

    2017

  15. [15]

    Yolo9000: Better, faster, stronger,

    J. Redmon and A. Farhadi, “Yolo9000: Better, faster, stronger,” in2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 6517–6525

    2017

  16. [16]

    Ffca-yolo for small object detection in remote sensing images,

    Y . Zhang, M. Ye, G. Zhu, Y . Liu, P. Guo, and J. Yan, “Ffca-yolo for small object detection in remote sensing images,”IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1–15, 2024

    2024

  17. [17]

    Lsodnet: A lightweight and efficient detector for small object detection in remote sensing images,

    K. Jin, W. Du, M. Tang, W. Liang, K. Li, and A.-S. K. Pathan, “Lsodnet: A lightweight and efficient detector for small object detection in remote sensing images,”IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 18, pp. 24 816–24 828, 2025

    2025

  18. [18]

    Drone-detr: Efficient small object detection for remote sensing image using enhanced rt-detr model,

    Y . Kong, X. Shang, and S. Jia, “Drone-detr: Efficient small object detection for remote sensing image using enhanced rt-detr model,” Sensors, vol. 24, no. 17, p. 5496, 2024

    2024

  19. [19]

    Rs-detr: An improved remote sensing object detection model based on rt-detr,

    J. Liet al., “Rs-detr: An improved remote sensing object detection model based on rt-detr,”Applied Sciences, vol. 14, no. 22, p. 10331, 2024

    2024

  20. [20]

    Transformers in small object detection: A benchmark and survey of state-of-the-art,

    A. Miri Rekavandi, S. Rashidi, F. Boussaid, S. Hoefs, E. Akbas, and M. Bennamoun, “Transformers in small object detection: A benchmark and survey of state-of-the-art,”ACM Computing Surveys, vol. 57, no. 4, pp. 1–37, 2024

    2024

  21. [21]

    Learning rotation-invariant convolu- tional neural networks for object detection in vhr optical remote sensing images,

    G. Cheng, P. Zhou, and J. Han, “Learning rotation-invariant convolu- tional neural networks for object detection in vhr optical remote sensing images,”IEEE transactions on geoscience and remote sensing, vol. 54, no. 12, pp. 7405–7415, 2016

    2016

  22. [22]

    Dota: A large-scale dataset for object detection in aerial images,

    G.-S. Xia, X. Bai, J. Ding, Z. Zhu, S. Belongie, J. Luo, M. Datcu, M. Pelillo, and L. Zhang, “Dota: A large-scale dataset for object detection in aerial images,” inProceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 3974–3983

    2018

  23. [23]

    Hierarchical and robust convolutional neural network for very high-resolution remote sensing object detection,

    Y . Zhang, Y . Yuan, Y . Feng, and X. Lu, “Hierarchical and robust convolutional neural network for very high-resolution remote sensing object detection,”IEEE Transactions on Geoscience and Remote Sens- ing, vol. 57, no. 8, pp. 5535–5548, 2019

    2019

  24. [24]

    Image super-resolution using deep convolutional networks,

    C. Dong, C. C. Loy, K. He, and X. Tang, “Image super-resolution using deep convolutional networks,”IEEE transactions on pattern analysis and machine intelligence, vol. 38, no. 2, pp. 295–307, 2015

    2015

  25. [25]

    Super-resolution for remote sensing images via local–global combined network,

    S. Lei, Z. Shi, and Z. Zou, “Super-resolution for remote sensing images via local–global combined network,”IEEE Geoscience and Remote Sensing Letters, vol. 14, no. 8, pp. 1243–1247, 2017

    2017

  26. [26]

    Deep distillation recursive network for remote sensing imagery super-resolution,

    K. Jiang, Z. Wang, P. Yi, J. Jiang, J. Xiao, and Y . Yao, “Deep distillation recursive network for remote sensing imagery super-resolution,”Remote Sensing, vol. 10, no. 11, p. 1700, 2018

    2018

  27. [27]

    Edge-enhanced gan for remote sensing image superresolution,

    K. Jiang, Z. Wang, P. Yi, G. Wang, T. Lu, and J. Jiang, “Edge-enhanced gan for remote sensing image superresolution,”IEEE Transactions on Geoscience and Remote Sensing, vol. 57, no. 8, pp. 5799–5812, 2019

    2019

  28. [28]

    Hierarchical dense recursive network for image super-resolution,

    K. Jiang, Z. Wang, P. Yi, and J. Jiang, “Hierarchical dense recursive network for image super-resolution,”Pattern Recognition, vol. 107, p. 107475, 2020

    2020

  29. [29]

    From degrade to upgrade: Learning a self-supervised degradation guided adaptive network for blind remote sensing image super-resolution,

    Y . Xiao, Q. Yuan, K. Jiang, J. He, Y . Wang, and L. Zhang, “From degrade to upgrade: Learning a self-supervised degradation guided adaptive network for blind remote sensing image super-resolution,”Information Fusion, vol. 96, pp. 297–311, 2023

    2023

  30. [30]

    Image super- resolution using very deep residual channel attention networks,

    Y . Zhang, K. Li, K. Li, L. Wang, B. Zhong, and Y . Fu, “Image super- resolution using very deep residual channel attention networks,” in Proceedings of the European conference on computer vision (ECCV), 2018, pp. 286–301

    2018

  31. [31]

    Single image super-resolution via a holistic attention network,

    B. Niu, W. Wen, W. Ren, X. Zhang, L. Yang, S. Wang, K. Zhang, X. Cao, and H. Shen, “Single image super-resolution via a holistic attention network,” inEuropean conference on computer vision. Springer, 2020, pp. 191–207

    2020

  32. [32]

    Hybrid-scale self-similarity exploitation for remote sensing image super-resolution,

    S. Lei and Z. Shi, “Hybrid-scale self-similarity exploitation for remote sensing image super-resolution,”IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1–10, 2022

    2022

  33. [33]

    Image super-resolution with non- local sparse attention,

    Y . Mei, Y . Fan, and Y . Zhou, “Image super-resolution with non- local sparse attention,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 3517–3526

    2021

  34. [34]

    Task-agnostic vision transformer for distributed learning of image processing,

    B. Kim, J. Kim, and J. C. Ye, “Task-agnostic vision transformer for distributed learning of image processing,”IEEE Transactions on Image Processing, vol. 32, pp. 203–218, 2022

    2022

  35. [35]

    Dster: A dense spectral transformer for remote sensing spectral super-resolution,

    J. He, Q. Yuan, J. Li, Y . Xiao, X. Liu, and Y . Zou, “Dster: A dense spectral transformer for remote sensing spectral super-resolution,”In- ternational Journal of Applied Earth Observation and Geoinformation, vol. 109, p. 102773, 2022

    2022

  36. [36]

    Local-global temporal difference learning for satellite video super- resolution,

    Y . Xiao, Q. Yuan, K. Jiang, X. Jin, J. He, L. Zhang, and C.-w. Lin, “Local-global temporal difference learning for satellite video super- resolution,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 34, no. 4, pp. 2789–2802, 2023

    2023

  37. [37]

    Swinir: Image restoration using swin transformer,

    J. Liang, J. Cao, G. Sun, K. Zhang, L. Van Gool, and R. Timofte, “Swinir: Image restoration using swin transformer,” inProceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 1833–1844

    2021

  38. [38]

    Ttst: A top- k token selective transformer for remote sensing image super-resolution,

    Y . Xiao, Q. Yuan, K. Jiang, J. He, C.-W. Lin, and L. Zhang, “Ttst: A top- k token selective transformer for remote sensing image super-resolution,” IEEE Transactions on Image Processing, vol. 33, pp. 738–752, 2024

    2024

  39. [39]

    Efficient swin transformer for remote sensing image super-resolution,

    X. Kang, P. Duan, J. Li, and S. Li, “Efficient swin transformer for remote sensing image super-resolution,”IEEE Transactions on Image Processing, vol. 33, pp. 6367–6379, 2024

    2024

  40. [40]

    Multi-scale object detection in remote sensing imagery with convolutional neural networks,

    Z. Deng, H. Sun, S. Zhou, J. Zhao, L. Lei, and H. Zou, “Multi-scale object detection in remote sensing imagery with convolutional neural networks,”ISPRS Journal of Photogrammetry and Remote Sensing, vol. 145, pp. 3–22, 2018

    2018

  41. [41]

    Fsanet: Feature-and-spatial-aligned network for tiny object detection in remote sensing images,

    J. Wu, Z. Pan, B. Lei, and Y . Hu, “Fsanet: Feature-and-spatial-aligned network for tiny object detection in remote sensing images,”IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1–17, 2022

    2022

  42. [42]

    Feature-enhanced centernet for small object detection in remote sensing images,

    T. Shi, J. Gong, J. Hu, X. Zhi, W. Zhang, Y . Zhang, P. Zhang, and G. Bao, “Feature-enhanced centernet for small object detection in remote sensing images,”Remote Sensing, vol. 14, no. 21, p. 5488, 2022

    2022

  43. [43]

    Legnet: A lightweight edge-gaussian network for low-quality remote sensing image object detection,

    W. Lu, S.-B. Chen, H.-D. Li, Q.-L. Shu, C. H. Ding, J. Tang, and B. Luo, “Legnet: A lightweight edge-gaussian network for low-quality remote sensing image object detection,” inProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2025, pp. 2844–2853

    2025

  44. [44]

    Pr-deformable detr: Detr for remote sensing object detection,

    Y . Chen, B. Liu, and L. Yuan, “Pr-deformable detr: Detr for remote sensing object detection,”IEEE Geoscience and Remote Sensing Letters, vol. 21, pp. 1–5, 2024

    2024

  45. [45]

    The effects of super-resolution on object detection performance in satellite imagery,

    J. Shermeyer and A. Van Etten, “The effects of super-resolution on object detection performance in satellite imagery,” in2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2019, pp. 1432–1441

    2019

  46. [46]

    Vehicle detection in remote sensing images leveraging on simultaneous super-resolution,

    H. Ji, Z. Gao, T. Mei, and B. Ramesh, “Vehicle detection in remote sensing images leveraging on simultaneous super-resolution,”IEEE Geoscience and Remote Sensing Letters, vol. 17, no. 4, pp. 676–680, 2019

    2019

  47. [47]

    From point to region: Accurate and efficient hierarchical small object detection in low-resolution remote sensing images,

    J. Wu and S. Xu, “From point to region: Accurate and efficient hierarchical small object detection in low-resolution remote sensing images,”Remote Sensing, vol. 13, no. 13, p. 2620, 2021

    2021

  48. [48]

    Dino: Detr with improved denoising anchor boxes for end-to- end object detection,

    H. Zhang, F. Li, S. Liu, L. Zhang, H. Su, J. Zhu, L. M. Ni, and H.-Y . Shum, “Dino: Detr with improved denoising anchor boxes for end-to- end object detection,” 2022

    2022

  49. [49]

    Swin transformer: Hierarchical vision transformer using shifted windows,

    Z. Liu, Y . Lin, Y . Cao, H. Hu, Y . Wei, Z. Zhang, S. Lin, and B. Guo, “Swin transformer: Hierarchical vision transformer using shifted windows,” inProceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 10 012–10 022

    2021

  50. [50]

    U-net: Convolutional networks for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” inInternational Conference on Medical image computing and computer-assisted intervention. Springer, 2015, pp. 234–241

    2015

  51. [51]

    Focal loss for dense object detection,

    T.-Y . Lin, P. Goyal, R. Girshick, K. He, and P. Doll ´ar, “Focal loss for dense object detection,” inProceedings of the IEEE international conference on computer vision, 2017, pp. 2980–2988

    2017

  52. [52]

    Cascade r-cnn: Delving into high quality object detection,

    Z. Cai and N. Vasconcelos, “Cascade r-cnn: Delving into high quality object detection,” inProceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 6154–6162

    2018

  53. [53]

    Yolox: Exceeding yolo series in 2021,

    Z. Ge, S. Liu, F. Wang, Z. Li, and J. Sun, “Yolox: Exceeding yolo series in 2021,”arXiv preprint arXiv:2107.08430, 2021

    Pith/arXiv arXiv 2021

  54. [54]

    Sparse r-cnn: An end-to-end framework for object detection,

    P. Sun, R. Zhang, Y . Jiang, T. Kong, C. Xu, W. Zhan, M. Tomizuka, Z. Yuan, and P. Luo, “Sparse r-cnn: An end-to-end framework for object detection,”IEEE transactions on pattern analysis and machine intelligence, vol. 45, no. 12, pp. 15 650–15 664, 2023

    2023

  55. [55]

    Fsanet: Feature-and-spatial-aligned network for tiny object detection in remote sensing images,

    J. Wu, Z. Pan, B. Lei, and Y . Hu, “Fsanet: Feature-and-spatial-aligned network for tiny object detection in remote sensing images,”IEEE Trans. Geosci. Remote Sens., vol. 60, pp. 1–17, 2022

    2022

  56. [56]

    Superyolo: Super resolution assisted object detection in multimodal remote sensing im- agery,

    J. Zhang, J. Lei, W. Xie, Z. Fang, Y . Li, and Q. Du, “Superyolo: Super resolution assisted object detection in multimodal remote sensing im- agery,”IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1–15, 2023

    2023

This paper was first reviewed by deepseek-v4-flash on August 3, 2026.