Pith. sign in

REVIEW 3 major objections 5 minor 80 references

DRRNet: Macro-Micro Feature Fusion and Dual Reverse Refinement for Camouflaged Object Detection

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read DRRNet pairs global and local feature fusion with two-stage reverse refinement to claim top scores on CAMO, COD10K, and NC4K.

desk verdict A coherent COD architecture paper whose headline SOTA claim is not backed by the reported numbers; worth referee time as a methods contribution, but the empirical superiority statement needs to be softened and re-supported. read the letter →

arxiv 2505.09168 v1 pith:PYK5LL5N submitted 2025-05-14 cs.CV cs.AI

Jianlin Sun , Xiaolin Fang , Juwei Guan , Dongdong Gui , Teqi Wang , Tongxin Zhu This is my paper · ORCID
classification cs.CVcs.AI
keywords camouflagedobjectdetectionfeaturefusionfrequency-domaincalibrationreverserefinementedge-awaresegmentationpyramidvisiontransformerspatial-frequencymodelingpolyp
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper introduces DRRNet, a four-stage network for camouflaged object detection built around a context-detail-fusion-refinement pipeline. It claims that separating global scene understanding (macro) from local microstructure cues (micro), fusing them in both spatial and frequency domains, and then applying two successive rounds of reverse refinement to a coarse prediction produces masks with stronger boundary continuity and less background interference. On the three standard benchmarks (CAMO, COD10K, NC4K), DRRNet reports the best or tied-best score on nearly every metric among 17 compared methods, without post-processing. A transfer experiment to polyp segmentation suggests the design generalizes to other low-contrast segmentation tasks. The practical payoff would be a single architecture that handles both the 'lost hair-like detail' failure of global-only models and the 'vegetation-pattern distraction' failure of local-only models.

What carries the argument

The load-bearing mechanism is the Dual Reverse Refinement Module (DRRM), applied twice in the decoder. Given coarse predictions from deeper stages $O_{i+1}$ and $O_{i+2}$, it forms inverted residual maps $R_1 = 1-\sigma(O_{i+1})$ and $R_2 = 1-\sigma(O_{i+2})$, uses their sum to weight the fused feature map, and adds the coarse outputs back as residual connections. In parallel, a frequency branch takes the Fourier transform of the fused features, modulates the spectrum with convolution-derived weights, and transforms back to the spatial domain. This dual spatial-frequency calibration, repeated in two stages, is what the paper credits for suppressing background noise while preserving boundary continuity.

What would settle it

An independent rerun of CAMO, COD10K, and NC4K using each competitor's official released prediction maps or official inference code, with the metric implementation fixed, would settle the claim: if EFNet, MVGNet, or BDCL-Net then matches or exceeds DRRNet on S-alpha or weighted F-measure, the reported state-of-the-art margin disappears. The paper's own conclusion names dynamic backgrounds and overlapping multi-target instances as remaining weak spots, so a stress benchmark built from video frames or densely overlapping camouflaged objects offers a second disconfirmation test.

Watch

Extended reading notes

Core claim

The central claim is that the longstanding COD trade-off between global context and local detail can be resolved architecturally rather than by choosing one side: DRRNet obtains panoramic context through a multi-branch Omni-Context Module, recovers microstructural cues through an ASPP-style and depthwise-separable Micro-Detail Module, and merges the two streams in a Macro-Micro Fusion module that jointly models spatial and frequency representations. The decoder first produces a coarse global prediction, then a Dual Reverse Refinement Module applies two inverse-refinement rounds, each using spatial edge priors from inverted sigmoid maps and frequency-domain noise suppression, to progressively remove background false positives and sharpen boundaries. Supported by ablations in which each module is removed or replaced, the paper argues that the components contribute additively and that the two-stage reverse refinement is what converts a rough localization into a precise segmentation. The paper's stated conclusion is that DRRNet outperforms all 17 compared state-of-the-art methods on CAMO, COD10K, and NC4K across S-measure, E-measure, weighted F-measure, and MAE without any post-processing.

Load-bearing premise

The strongest empirical claim rests on the fairness of the comparisons: the paper states that all prediction maps were generated with identical code, and if the reproduced baselines were not optimally tuned or their official inference code was not used, the few-thousandths-of-a-point gains in Table I are not meaningful.

Editorial extensions

If this is right

  • If the reported numbers hold, DRRNet becomes the new reference point on CAMO, COD10K, and NC4K, with the highest weighted F-measure on COD10K (0.817) and the lowest MAE (0.019) among the compared methods.
  • The two-stage reverse refinement recipe—coarse decode, then invert and re-weight with edge priors—can be lifted out and attached to other encoder-decoder detectors as a plug-in decoder.
  • The spatial-plus-frequency fusion pattern gives later COD designs a concrete alternative to stacking more Transformer blocks: the gains here come from explicit dual-domain calibration rather than larger receptive fields.
  • Because the same network transfers to polyp segmentation with only retraining, low-contrast medical segmentation tasks are a direct application area if the COD results replicate.

Reading between the lines

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

  • The paper does not run a direct comparison against the officially released prediction maps of its closest competitors; a natural test is to re-evaluate on those maps to see whether the roughly 0.001-point S-alpha gaps persist or shrink to noise.
  • The paper's own closing caveat—dynamic backgrounds and overlapping multi-target instances remain hard—suggests the next stress test should be a video or multi-instance camouflage benchmark, where the current single-image metrics may not reflect the claimed robustness.
  • Because the ablation tables show metric gains of only a few thousandths on some settings, an independent rerun with multiple random seeds would clarify which modules carry the real signal; the paper reports single-run comparisons.
  • The frequency-domain mechanism works by modulating Fourier amplitude coefficients, so a plausible inference is that high-frequency texture noise in vegetation-like backgrounds is the main source of false positives DRRNet suppresses; this could be tested by adding synthetic high-frequency noise to test images and measuring MAE degradation.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The manuscript proposes DRRNet, a four-stage camouflaged object detection (COD) architecture built around an Omni-Context Module (OCM) for global context, a Micro Detail Module (MDM) for local details, a Macro-Micro Fusion (MMF) module for spatial-frequency fusion, a Global Rough Decoder (GRD), and a Dual Reverse Refinement Module (DRRM) that performs two rounds of spatial-edge and frequency-domain refinement. The method is evaluated on CAMO, COD10K, and NC4K against 17 published methods, with additional experiments on four polyp segmentation datasets. The authors report that DRRNet outperforms all compared state-of-the-art methods on nearly all metrics, and they provide ablation studies for each architectural component. The central claim is that DRRNet 'significantly outperforms all SOTA methods across all datasets' without post-processing.

Significance. If the reported numbers are reproducible and the comparisons are fair, DRRNet would be a competitive COD architecture with a clear and well-motivated design. The paper follows standard COD evaluation practice, uses three widely adopted benchmarks and four metrics, ablates every proposed module, reports parameter and FLOP counts, and releases code. The modular decomposition into global-context, local-detail, fusion, and reverse-refinement components is sensible, and the polyp segmentation transfer is a useful generalization check. The principal weakness is that the headline claim of significant state-of-the-art improvement rests on point-estimate differences of 0.001–0.004 in several metrics, with exact ties on two key entries, and no variance, confidence intervals, or paired significance tests are reported. As a result, the empirical superiority clause is not currently established, even though the architecture itself appears sound and the ablations support the qualitative contribution of each module.

major comments (3)
  1. [Section IV.B, Table I] The claim that 'DRRNet significantly outperforms all SOTA methods across all datasets' is not supported by the reported evidence. In Table I, the largest margins over the closest competitor are typically 0.001–0.004 in Sα and Eϕ, and there are exact ties: CAMO Sα is 0.881 for both DRRNet and EFNet, and NC4K Sα is 0.896 for both. No confidence intervals, standard deviations across training seeds, or paired statistical tests are provided. Since COD training is stochastic and run-to-run variation in Sα is typically larger than these margins, the reported numbers are consistent with the null hypothesis that DRRNet is not better than EFNet or BDCL-Net. The authors should add multi-seed results with variance reporting or paired significance tests across the test sets, or alternatively soften the claim to describe the observed point estimates.
  2. [Section IV.B] The fairness statement 'all prediction maps are generated using identical code, and all metrics are computed with the same implementation' is ambiguous and does not establish that the 17 baselines were not disadvantaged. It is not stated whether each baseline used its official released checkpoint and official inference code, or whether all baseline predictions were regenerated under a unified pipeline that may not match each model's original inference procedure. If any baseline was re-implemented, re-trained, or run with non-official weights, its Table I numbers could be depressed by more than the claimed advantage. The authors should specify, for every baseline, the source of the predictions (official checkpoint and code vs. re-implementation), and report the exact protocol used. This is load-bearing because the central claim of SOTA improvement depends on the fairness of these comparisons.
  3. [Section IV.D, Table VII] The polyp segmentation extension is presented as evidence of generalization, but the claim that DRRNet 'significantly outperforms existing methods' is again based on single-run point estimates with no variance or significance testing. For example, on CVC-300 the reported MAE improvement over GLCONet is 0.009 to 0.007, and on Kvasir the Sm gap is 0.925 vs. 0.872 for GLCONet; these are large in some cases, but no error bars or statistical tests are provided. This is a secondary claim relative to the COD result, but it should be reported with the same caveats or supported by appropriate statistics.
minor comments (5)
  1. [Section III.B, Eq. (5)] Equation (5) has a mismatched parenthesis: the expression 'li = CBR(Φ(SE(cat(Faspp,Fdw),Fres))+x0' appears to be missing a closing parenthesis after the CBR argument. Please fix the equation for clarity.
  2. [Section III.C, Eq. (6)] In Eq. (6), the spatial branch is written as 'xspatial = CBR(Φ(gi,li))', but the textual description says the GroupFusionBlock operates on each channel group separately. This notational mismatch makes the actual grouping operation ambiguous. Also, the expression for xfreq is missing a closing parenthesis after the modulation term.
  3. [References, [32]] The text cites [32] as the Pyramid Vision Transformer (PVTv2) backbone, but the reference listed is an Applied Sciences paper titled 'Camouflaged Object Detection That Does Not Require Additional Priors.' This appears to be the wrong reference; the correct PVTv2 citation should be provided.
  4. [Section III.A] The sentence '...which significantly improving the accuracy of camouflaged object segmentation' contains a grammatical error; it should read '...which significantly improves the accuracy...'
  5. [Table II] The checkmark layout of Table II is difficult to read: the header includes 'Baseline' alongside the five module columns, and it is not immediately clear which row corresponds to the baseline and which row corresponds to the full model. Please use explicit row labels such as 'Baseline', '+OCM', '+MDM', etc., and state in the caption what each checkmark denotes.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: DRRNet's claims rest on standard training/test evaluation and ablations, not on fitted constants or self-citation chains.

full rationale

The paper's central claim is empirical: DRRNet outperforms SOTA on CAMO, COD10K, and NC4K. The derivation chain is a normal supervised-learning loop: PVTv2 features are fed through OCM, MDM, MMF, GRD, and DRRM; the network is trained with weighted BCE and IoU losses on the standard COD10K/CAMO training split; and the reported metrics are computed on held-out test partitions. No parameter is fitted to the test set and then renamed as a prediction, and no quantity in Eq. (9) or the module equations is defined in terms of the final Table I results. The self-citations to SDRNet [20] and EFNet [21] are used only as comparison baselines in Table I and Fig. 7, not as load-bearing justification for DRRNet's architecture or for excluding alternatives; citing one's own earlier baselines is normal and does not make the present evaluation circular. The design choices (multi-branch context, ASPP/depthwise detail extraction, frequency-domain modulation, reverse refinement) are heuristics validated by ablations in Tables II-V, which is a legitimate empirical support structure. Concerns that the Table I gains are within run-to-run noise, or that the Section IV.B fairness sentence does not prove official checkpoints were used, are statistical-validity concerns, not circularity; they do not make any predicted quantity equivalent to an input by construction. Therefore the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim is an empirical architecture result. The paper introduces no new physical or conceptual entities; the modules OCM, MDM, MMF, GRD, and DRRM are software components whose existence is the code itself. The free parameters are the hand-chosen training and architecture settings that the reported results depend on.

free parameters (3)
  • Training schedule hyperparameters = lr=1e-4, decay 0.1 per 25 epochs, batch size 8, 80 epochs, input 384x384
    Chosen by hand in Section IV.A; needed to reproduce the reported results and not derived from any principle.
  • Number of DRRM refinement stages = 2
    The 'dual' in Dual Reverse Refinement is a design choice; the paper does not test whether two stages are optimal versus one or three.
  • Module hyperparameters = dilation rates {1,3,5,7}, depthwise kernels {3,5,7}, number of MMF channel groups 4
    Selected arbitrarily and reported in Section III; the ablation studies compare module presence but not these specific values.
assumptions (4)
  • domain assumption PVTv2 pretrained on ImageNet provides suitable multi-scale backbone features for COD.
    Invoked in Section III.A and Implementation Details; the entire architecture is built on this backbone without comparison to alternative backbones.
  • domain assumption The standard COD datasets (CAMO, COD10K, NC4K) and metrics (S-alpha, E-phi, F-beta-w, MAE) are the accepted measure of camouflaged object detection quality.
    Used throughout Section IV; the significance of the results depends on these benchmarks being representative and the metrics being reliable.
  • domain assumption The compared SOTA methods were fairly reproduced using identical code and the same metric implementation.
    Stated in Section IV.B: 'all prediction maps are generated using identical code'; the main comparative claim rests on this fairness premise.
  • domain assumption Supervised training with weighted BCE and IoU losses, as in Equation 9, converges to a model that generalizes to the withheld test images.
    Assumed in the experimental setup; no analysis of convergence or overfitting is provided beyond final test-set numbers.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DRRNet: Macro-Micro Feature Fusion and Dual Reverse Refinement for Camouflaged Object Detection." pith.science (2026). https://pith.science/paper/PYK5LL5N

@misc{pith2026250509168,
  author       = {Pith},
  title        = {Pith review of: DRRNet: Macro-Micro Feature Fusion and Dual Reverse Refinement for Camouflaged Object Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PYK5LL5N}},
  note         = {Machine review of arXiv:2505.09168}
}
read the original abstract

The core challenge in Camouflage Object Detection (COD) lies in the indistinguishable similarity between targets and backgrounds in terms of color, texture, and shape. This causes existing methods to either lose edge details (such as hair-like fine structures) due to over-reliance on global semantic information or be disturbed by similar backgrounds (such as vegetation patterns) when relying solely on local features. We propose DRRNet, a four-stage architecture characterized by a "context-detail-fusion-refinement" pipeline to address these issues. Specifically, we introduce an Omni-Context Feature Extraction Module to capture global camouflage patterns and a Local Detail Extraction Module to supplement microstructural information for the full-scene context module. We then design a module for forming dual representations of scene understanding and structural awareness, which fuses panoramic features and local features across various scales. In the decoder, we also introduce a reverse refinement module that leverages spatial edge priors and frequency-domain noise suppression to perform a two-stage inverse refinement of the output. By applying two successive rounds of inverse refinement, the model effectively suppresses background interference and enhances the continuity of object boundaries. Experimental results demonstrate that DRRNet significantly outperforms state-of-the-art methods on benchmark datasets. Our code is available at https://github.com/jerrySunning/DRRNet.

Figures

Figures reproduced from arXiv: 2505.09168 by the authors.

Figure 1
Figure 1. Visual comparison of models utilizing various types of [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The framework of the proposed DRRNet.OCM extracts panoramic contextual semantics, MDM extracts local detail [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Details of the proposed OmniContext Module (left) and MicroDetail Module (right). [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: The details of Macro-Micro Fusion(MMF) module. [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: The details of Global Rough Decoder(GRD) module. [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: The details of Dual Reverse Refinement Mod [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: The visualization comparisons between our DRRNet and other stronger SOTA methods in various scenarios. Under the [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Visual results of the effectiveness of our modules. [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: Visual results of the different prediction output of our model. [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: Qualitative visual comparisons of our DRRNet [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

80 extracted references · 39 canonical work pages

  1. [13]

    FAPNet: An Effective Frequency Adap- tive Point-based Eye Tracker,

    X. Lin, H. Ren, and B. Cheng, “FAPNet: An Effective Frequency Adap- tive Point-based Eye Tracker,” 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 5789–5798, Jun. 2024, doi: https://doi.org/10.1109/cvprw63382.2024.00588

    arXiv 2024

  2. [32]

    Cam- ouflaged Object Detection That Does Not Require Additional Pri- ors,

    Y . Dong, H. Zhou, C. Li, J. Xie, Y . Xie, and Z. Li, “Cam- ouflaged Object Detection That Does Not Require Additional Pri- ors,” Applied Sciences, vol. 14, no. 6, p. 2621, Jan. 2024, doi: https://doi.org/10.3390/app14062621

    work page doi:10.3390/app14062621 2024

  3. [55]

    Deep Gradient Learning for Efficient Camouflaged Object De- tection,

    G.-P. Ji, D.-P. Fan, Y .-C. Chou, D. Dai, A. Liniger, and L. V . Gool, “Deep Gradient Learning for Efficient Camouflaged Object De- tection,” Deleted Journal, vol. 20, no. 1, pp. 92–108, Jan. 2023, doi: https://doi.org/10.1007/s11633-022-1365-9

    work page doi:10.1007/s11633-022-1365-9 2023

  4. [1]

    Predator-specific camouflage in chameleons,

    D. Stuart-Fox, A. Moussalli, and M. J. Whiting, “Predator-specific camouflage in chameleons,” Biology Letters, vol. 4, no. 4, pp. 326–329, May 2008

    work page 2008

  5. [2]

    What, if anything, is the adaptive function of countershading?,

    G. D. Ruxton, M. P. Speed, and D. J. Kelly, “What, if anything, is the adaptive function of countershading?,” Animal Behaviour, vol. 68, no. 3, pp. 445–451, Sep. 2004

    work page 2004

  6. [3]

    Boundary- Guided Camouflaged Object Detection,

    Y . Sun, S. Wang, C. Chen, and T.-Z. Xiang, “Boundary- Guided Camouflaged Object Detection,” arXiv.org, 2022. https://arxiv.org/abs/2207.00794 (accessed May 09, 2025)

    arXiv 2022

  7. [4]

    CDDNet: Camouflaged Defect Detection Network for Steel Surface,

    Q. Luo et al., “CDDNet: Camouflaged Defect Detection Network for Steel Surface,” IEEE Transactions on Instrumentation and Measurement, vol. 73, pp. 1–13, 2024, doi: https://doi.org/10.1109/tim.2023.3336452

    arXiv 2024

  8. [5]

    Camouflage Defect Identification: A Novel Approach,

    N. U. Bhajantri and P. Nagabhushan, “Camouflage Defect Identification: A Novel Approach,” 9th International Conference on Information Tech- nology (ICIT’06), Dec. 2006, doi: https://doi.org/10.1109/icit.2006.34

    work page doi:10.1109/icit.2006.34 2006

Show all 80 references
  1. [6]

    SDDNet: A Fast and Accurate Network for Surface Defect Detection,

    L. Cui, X. Jiang, M. Xu, W. Li, P. Lv, and B. Zhou, “SDDNet: A Fast and Accurate Network for Surface Defect Detection,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1–13, Jan. 2021, doi: https://doi.org/10.1109/tim.2021.3056744

    2021

  2. [7]

    Synthetic data augmentation for surface defect detection and classification using deep learning,

    S. Jain, G. Seth, A. Paruthi, U. Soni, and G. Kumar, “Synthetic data augmentation for surface defect detection and classification using deep learning,” Journal of Intelligent Manufacturing, Nov. 2020, doi: https://doi.org/10.1007/s10845-020-01710-x

    2020 doi

  3. [8]

    Partially Camouflaged Object Tracking using Modified Probabilistic Neural Network and Fuzzy Energy based Active Contour,

    A. Mondal, S. Ghosh, and A. Ghosh, “Partially Camouflaged Object Tracking using Modified Probabilistic Neural Network and Fuzzy Energy based Active Contour,” International Journal of Computer Vision, vol. 122, no. 1, pp. 116–148, Oct. 2016, doi: https://doi.org/10.1007/s11263-...

    2016 doi

  4. [9]

    Texture segmen- tation by multiscale aggregation of filter responses and shape elements,

    None Galun, N. Sharon, None Basri, and N. Brandt, “Texture segmen- tation by multiscale aggregation of filter responses and shape elements,” CiteSeer X (The Pennsylvania State University), pp. 716–723 vol.1, Jan. 2003, doi: https://doi.org/10.1109/iccv.2003.1238418

    2003 arXiv

  5. [10]

    A Bayesian Approach to Camouflaged Moving Object Detection,

    X. Zhang, C. Zhu, Wang Shuai, Y . Liu, and M. Ye, “A Bayesian Approach to Camouflaged Moving Object Detection,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 27, no. 9, pp. 2001–2013, Sep. 2017, doi: https://doi.org/10.1109/tcsvt.2016.2555719

    2001

  6. [11]

    Fan, G.-P

    D.-P. Fan, G.-P. Ji, G. Sun, M.-M. Cheng, J. Shen, L. Shao, Camou- flaged object detection, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 2777–2787

    2020

  7. [12]

    Context-aware Cross- level Fusion Network for Camouflaged Object Detection,

    Y . Sun, G. Chen, T. Zhou, Y . Zhang, and N. Liu, “Context-aware Cross- level Fusion Network for Camouflaged Object Detection,” arXiv.org,

  8. [14]

    Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet,

    L. Yuan et al., “Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet,” arXiv:2101.11986 [cs], Mar. 2021, Available: https://arxiv.org/abs/2101.11986

    2021 arXiv

  9. [15]

    Visual Saliency Transformer,

    N. Liu, N. Zhang, K. Wan, L. Shao, and J. Han, “Visual Saliency Transformer,” arXiv.org, 2021. https://arxiv.org/abs/2104.12099 (accessed May 09, 2025)

    2021 arXiv

  10. [16]

    U-Net: Convolutional Net- works for Biomedical Image Segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Net- works for Biomedical Image Segmentation,” arXiv.org, May 18, 2015. https://arxiv.org/abs/1505.04597

    2015 arXiv

  11. [17]

    CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows

    X. Dong et al., “CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows.” Accessed: May 02, 2023. [On- line]. Available: https://arxiv.org/pdf/2107.00652.pdf

    2023 arXiv

  12. [18]

    Frequency-Spatial Entanglement Learning for Camouflaged Object Detection,

    Y . Sun, C. Xu, J. Yang, H. Xuan, and L. Luo, “Frequency-Spatial Entanglement Learning for Camouflaged Object Detection,” arXiv.org,

  13. [19]

    IdeNet: Making Neural Network Identify Camou- flaged Objects Like Creatures,

    “IdeNet: Making Neural Network Identify Camou- flaged Objects Like Creatures,” Ieee.org, 2021, doi: https://doi.org/10.1109/TIP.2024.3449574

    2021

  14. [20]

    SDRNet: Camouflaged object detection with independent reconstruction of structure and de- tail,

    J. Guan, X. Fang, T. Zhu, and W. Qian, “SDRNet: Camouflaged object detection with independent reconstruction of structure and de- tail,” Knowledge-Based Systems, vol. 299, p. 112051, Jun. 2024, doi: https://doi.org/10.1016/j.knosys.2024.112051

    2024

  15. [21]

    Promoting camouflaged object detection through novel edge–target interaction and frequency- spatial fusion,

    J. Guan, W. Qian, T. Zhu, and X. Fang, “Promoting camouflaged object detection through novel edge–target interaction and frequency- spatial fusion,” Neurocomputing, vol. 617, p. 129064, Dec. 2024, doi: https://doi.org/10.1016/j.neucom.2024.129064

    2024

  16. [22]

    Knowledge Rectification for Camouflaged Object De- tection: Unlocking Insights from Low-Quality Data,

    J. Guan et al., “Knowledge Rectification for Camouflaged Object De- tection: Unlocking Insights from Low-Quality Data,” arXiv.org, 2025. https://arxiv.org/abs/2503.22180 (accessed May 09, 2025)

    2025 arXiv

  17. [23]

    Anabranch network for camouflaged object segmentation,

    T.-N. Le, T. V . Nguyen, Z. Nie, M.-T. Tran, and A. Sugimoto, “Anabranch network for camouflaged object segmentation,” Computer Vision and Image Understanding, vol. 184, pp. 45–56, Jul. 2019, doi: https://doi.org/10.1016/j.cviu.2019.04.006

    2019 doi

  18. [24]

    Simultaneously Localize, Segment and Rank the Camouflaged Objects,

    Yunqiu Lv et al., “Simultaneously Localize, Segment and Rank the Camouflaged Objects,” arXiv (Cornell University), Jun. 2021, doi: https://doi.org/10.1109/cvpr46437.2021.01142

    2021

  19. [25]

    I Can Find You! Boundary-Guided Separated Attention Network for Camouflaged Object Detection,

    H. Zhu et al., “I Can Find You! Boundary-Guided Separated Attention Network for Camouflaged Object Detection,” Proceedings of the ... AAAI Conference on Artificial Intelligence, vol. 36, no. 3, pp. 3608–3616, Jun. 2022, doi: https://doi.org/10.1609/aaai.v36i3.20273

    2022 doi

  20. [26]

    MirrorNet: Bio-Inspired Camouflaged Object Segmentation,

    J. Yan, T.-N. Le, K.-D. Nguyen, M.-T. Tran, T.-T. Do, and T. V . Nguyen, “MirrorNet: Bio-Inspired Camouflaged Object Segmentation,” IEEE Access, vol. 9, pp. 43290–43300, 2021

    2021

  21. [27]

    Zoom In and Out: A Mixed-scale Triplet Network for Camouflaged Object Detection,

    Y . Pang, X. Zhao, T.-Z. Xiang, L. Zhang, and H. Lu, “Zoom In and Out: A Mixed-scale Triplet Network for Camouflaged Object Detection,” arXiv.org, 2022. https://arxiv.org/abs/2203.02688

    2022 arXiv

  22. [28]

    OS- Former: One-Stage Camouflaged Instance Segmentation with Transform- ers,

    J. Pei, T. Cheng, D.-P. Fan, H. Tang, C. Chen, and V . Gool, “OS- Former: One-Stage Camouflaged Instance Segmentation with Transform- ers,” arXiv.org, 2022. https://arxiv.org/abs/2207.02255 (accessed May 10, 2025)

    2022 arXiv

  23. [29]

    Improving Cam- ouflaged Object Detection with the Uncertainty of Pseudo-edge Labels,

    Nobukatsu Kajiura, H. Liu, and Shin’ichi Satoh, “Improving Cam- ouflaged Object Detection with the Uncertainty of Pseudo-edge Labels,” arXiv (Cornell University), pp. 1–7, Dec. 2021, doi: https://doi.org/10.1145/3469877.3490587

    2021

  24. [31]

    Semantic- Aware Domain Generalized Segmentation,

    D. Peng, Y . Lei, M. Hayat, Y . Guo, and W. Li, “Semantic- Aware Domain Generalized Segmentation,” arXiv.org, 2022. https://arxiv.org/abs/2204.00822 (accessed May 10, 2025)

    2022 arXiv

  25. [33]

    GLCONet: Learning Multi- source Perception Representation for Camouflaged Object Detection,

    Y . Sun, H. Xuan, J. Yang, and L. Luo, “GLCONet: Learning Multi- source Perception Representation for Camouflaged Object Detection,” arXiv.org, 2024. https://arxiv.org/abs/2409.09588 (accessed May 10, 2025)

    2024 arXiv

  26. [34]

    FOCUS: Towards Universal Foreground Segmentation,

    Z. You, L. Kong, L. Meng, and Z. Wu, “FOCUS: Towards Universal Foreground Segmentation,” arXiv.org, 2025. https://arxiv.org/abs/2501.05238 (accessed May 10, 2025)

    2025 arXiv

  27. [35]

    Squeeze-and- Excitation Networks,

    J. Hu, L. Shen, S. Albanie, G. Sun, and E. Wu, “Squeeze-and- Excitation Networks,” arXiv:1709.01507 [cs], May 2019, Available: https://arxiv.org/abs/1709.01507

    2019 arXiv

  28. [36]

    CBAM: Convolutional Block Attention Module,

    S. Woo, J. Park, J.-Y . Lee, and I. S. Kweon, “CBAM: Convolutional Block Attention Module,” arXiv:1807.06521 [cs], Jul. 2018, Available: https://arxiv.org/abs/1807.06521

    2018 arXiv

  29. [37]

    Deep Residual Learning for Image Recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” arXiv.org, Dec. 10, 2015. https://arxiv.org/abs/1512.03385

    2015 arXiv

  30. [38]

    DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs,

    L.-C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille, “DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 40, no. 4, pp. 834–848, ...

    2018

  31. [39]

    Fusion- Painting: Multimodal Fusion with Adaptive Attention for 3D Object Detection,

    S. Xu, D. Zhou, J. Fang, J. Yin, Z. Bin, and L. Zhang, “Fusion- Painting: Multimodal Fusion with Adaptive Attention for 3D Object Detection,” arXiv.org, 2021. https://arxiv.org/abs/2106.12449 (accessed May 10, 2025)

    2021 arXiv

  32. [40]

    Free-Form Image Inpainting with Gated Convolution,

    J. Yu, Z. Lin, J. Yang, X. Shen, X. Lu, and T. Huang, “Free-Form Image Inpainting with Gated Convolution,” arXiv:1806.03589 [cs], Oct. 2019, Available: https://arxiv.org/abs/1806.03589

    2019 arXiv

  33. [41]

    FMambaIR: A Hybrid State-Space Model and Frequency Domain for Image Restoration,

    X. Luan et al., “FMambaIR: A Hybrid State-Space Model and Frequency Domain for Image Restoration,” IEEE Transactions on Geoscience and Remote Sensing, vol. 63, pp. 1–14, 2025, doi: https://doi.org/10.1109/tgrs.2025.3526927

    2025

  34. [42]

    The fast Fourier trans- form,

    E. O. Brigham and R. E. Morrow, “The fast Fourier trans- form,” IEEE Spectrum, vol. 4, no. 12, pp. 63–70, Dec. 1967, doi: https://doi.org/10.1109/mspec.1967.5217220

    1967

  35. [43]

    MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,

    A. G. Howard et al., “MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,” arXiv.org, Apr. 17, 2017. https://arxiv.org/abs/1704.04861

    2017 arXiv

  36. [44]

    Gaussian Error Linear Units (GELUs),

    D. Hendrycks and K. Gimpel, “Gaussian Error Linear Units (GELUs),” arXiv:1606.08415 [cs], Jul. 2020, Available: https://arxiv.org/abs/1606.08415

    2020 arXiv

  37. [45]

    Feature Shrinkage Pyramid for Camouflaged Object Detection with Transformers

    Z. Huang et al., “Feature Shrinkage Pyramid for Camouflaged Object Detection with Transformers.” Accessed: May 10, 2025. [Online]

    2025

  38. [46]

    Feature Aggregation and Propagation Network for Camouflaged Object Detection,

    T. Zhou, Y . Zhou, C. Gong, J. Yang, and Y . Zhang, “Feature Aggregation and Propagation Network for Camouflaged Object Detection,” IEEE transactions on image processing, vol. 31, pp. 7036–7047, Jan. 2022, doi: https://doi.org/10.1109/tip.2022.3217695

    2022

  39. [47]

    Optimizing Intersection-Over-Union in Deep Neural Networks for Image Segmentation,

    M. A. Rahman and Y . Wang, “Optimizing Intersection-Over-Union in Deep Neural Networks for Image Segmentation,” Advances in Visual Computing, pp. 234–244, 2016

    2016

  40. [48]

    Structure- Measure: A New Way to Evaluate Foreground Maps,

    D.-P. Fan, M.-M. Cheng, Y . Liu, T. Li, and A. Borji, “Structure- Measure: A New Way to Evaluate Foreground Maps,” 2017 IEEE International Conference on Computer Vision (ICCV), Oct. 2017, doi: https://doi.org/10.1109/iccv.2017.487

    2017 doi

  41. [49]

    Enhanced-alignment Measure for Binary Foreground Map Evaluation

    D.-P. Fan, C. Gong, Y . Cao, B. Ren, M.-M. Cheng, and A. Borji, “Enhanced-alignment Measure for Binary Foreground Map Evaluation.” Accessed: May 10, 2025. [Online]. Available: https://www.ijcai.org/Proceedings/2018/0097.pdf

    2025

  42. [50]

    How to Evaluate Foreground Maps,

    R. Margolin, L. Zelnik-Manor, and A. Tal, “How to Evaluate Foreground Maps,” IEEE Xplore, Jun. 01, 2014. https://ieeexplore.ieee.org/document/6909433 (accessed Mar. 11, 2021)

    2014

  43. [51]

    Im- ageNet: A large-scale hierarchical image database,

    J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, “Im- ageNet: A large-scale hierarchical image database,” 2009 IEEE Con- ference on Computer Vision and Pattern Recognition, Jun. 2009, doi: https://doi.org/10.1109/cvpr.2009.5206848

    2009

  44. [52]

    Adam: A Method for Stochastic Optimization,

    D. P. Kingma and J. Ba, “Adam: A Method for Stochastic Optimization,” arXiv.org, Dec. 22, 2014. https://arxiv.org/abs/1412.6980

    2014 arXiv

  45. [53]

    Camouflaged Object Detection with Feature Decom- position and Edge Reconstruction

    [1]C. He et al., “Camouflaged Object Detection with Feature Decom- position and Edge Reconstruction.” Accessed: May 10, 2025. [Online]

    2025

  46. [54]

    Edge-Aware Mirror Net- work for Camouflaged Object Detection,

    D. Sun, S. Jiang, and L. Qi, “Edge-Aware Mirror Net- work for Camouflaged Object Detection,” arXiv.org, 2023. https://arxiv.org/abs/2307.03932 (accessed May 10, 2025)

    2023 arXiv

  47. [56]

    Locate, enhance and fuse: a progressively optimized network for camouflaged object detection,

    T. Qiu et al., “Locate, enhance and fuse: a progressively optimized network for camouflaged object detection,” Multimedia Tools and Ap- plications, Jul. 2024, doi: https://doi.org/10.1007/s11042-024-19891-5

    2024 doi

  48. [57]

    Camouflaged Object Segmentation with Omni Percep- tion,

    H. Mei et al., “Camouflaged Object Segmentation with Omni Percep- tion,” International Journal of Computer Vision, vol. 131, no. 11, pp. 3019–3034, Jul. 2023, doi: https://doi.org/10.1007/s11263-023-01838-2

    2023 doi

  49. [58]

    UEDG:Uncertainty-Edge Dual Guided Camouflage Object Detection,

    Y . Lyu, H. Zhang, Y . Li, H. Liu, Y . Yang, and D. Yuan, “UEDG:Uncertainty-Edge Dual Guided Camouflage Object Detection,” IEEE Transactions on Multimedia, vol. 26, pp. 4050–4060, 2024, doi: https://doi.org/10.1109/tmm.2023.3295095

    2024

  50. [59]

    MSCAF-Net: A Gen- eral Framework for Camouflaged Object Detection via Learning Multi-Scale Context-Aware Features,

    Y . Liu, H. Li, J. Cheng, and X. Chen, “MSCAF-Net: A Gen- eral Framework for Camouflaged Object Detection via Learning Multi-Scale Context-Aware Features,” IEEE Transactions on Cir- cuits and Systems for Video Technology, pp. 1–1, 2023, doi: https://doi.org/10.1109/tcsvt.2023.3245883

    2023

  51. [60]

    Bilateral decoupling complementarity learning network for camouflaged object detection,

    R. Zhao, Y . Li, Q. Zhang, and X. Zhao, “Bilateral decoupling complementarity learning network for camouflaged object detection,” Knowledge-Based Systems, vol. 314, p. 113158, Feb. 2025, doi: https://doi.org/10.1016/j.knosys.2025.113158

    2025

  52. [61]

    A Universal Multi- View Guided Network for Salient Object and Camouflaged Object Detection,

    X. Song, P. Zhang, X. Lu, X. Hei, and R. Liu, “A Universal Multi- View Guided Network for Salient Object and Camouflaged Object Detection,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 34, no. 11, pp. 11184–11197, Nov. 2024, doi: https://doi.org/10.1109...

    2024

  53. [62]

    Learning Discriminative Representations From Cross-Scale Features for Camouflaged Object Detection,

    Y . Wang, X. Bi, B. Liu, Y . Wei, W. Li, and B. Xiao, “Learning Discriminative Representations From Cross-Scale Features for Camouflaged Object Detection,” IEEE Transactions on Cir- 13 cuits and Systems for Video Technology, pp. 1–1, Jan. 2024, doi: https://doi.org/10.1109/tcs...

    2024

  54. [63]

    Finding Camouflaged Objects Along the Camouflage Mechanisms,

    Y . Yang and Q. Zhang, “Finding Camouflaged Objects Along the Camouflage Mechanisms,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 34, no. 4, pp. 2346–2360, Aug. 2023, doi: https://doi.org/10.1109/tcsvt.2023.3308964

    2023

  55. [64]

    FLRNet: A bio-inspired three- stage network for Camouflaged Object Detection via filtering, local- ization and refinement,

    Y . Zhao, Q. Zhang, and Y . Li, “FLRNet: A bio-inspired three- stage network for Camouflaged Object Detection via filtering, local- ization and refinement,” Neurocomputing, p. 129523, Feb. 2025, doi: https://doi.org/10.1016/j.neucom.2025.129523

    2025

  56. [65]

    Progressive Region-to-Boundary Exploration Network for Camouflaged Object Detection,

    G. Yue et al., “Progressive Region-to-Boundary Exploration Network for Camouflaged Object Detection,” IEEE Transactions on Multimedia, vol. 27, pp. 236–248, 2025, doi: https://doi.org/10.1109/tmm.2024.3521761

    2025

  57. [66]

    Adaptive Guid- ance Learning for Camouflaged Object Detection,

    Z. Chen, X. Zhang, T.-Z. Xiang, and Y . Tai, “Adaptive Guid- ance Learning for Camouflaged Object Detection,” arXiv.org, 2024. https://arxiv.org/abs/2405.02824 (accessed May 10, 2025)

    2024 arXiv

  58. [67]

    Camou- flaged Object Segmentation with Distraction Mining,

    H. Mei, G.-P. Ji, Z. Wei, X. Yang, X. Wei, and D.-P. Fan, “Camou- flaged Object Segmentation with Distraction Mining,” arXiv.org, 2021. https://arxiv.org/abs/2104.10475 (accessed May 10, 2025)

    2021 arXiv

  59. [68]

    Mutual Graph Learning for Camouflaged Object Detection,

    Q. Zhai, X. Li, F. Yang, C. Chen, H. Cheng, and D.-P. Fan, “Mutual Graph Learning for Camouflaged Object Detection,” arXiv.org, 2021. https://arxiv.org/abs/2104.02613 (accessed May 10, 2025)

    2021 arXiv

  60. [69]

    Uncertainty-Guided Transformer Reasoning for Camouflaged Object Detection,

    F. Yang et al., “Uncertainty-Guided Transformer Reasoning for Camouflaged Object Detection,” 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Oct. 2021, doi: https://doi.org/10.1109/iccv48922.2021.00411

    2021

  61. [70]

    Uncertainty-aware Joint Salient Object and Camouflaged Object Detection,

    A. Li, J. Zhang, Y . Lv, B. Liu, T. Zhang, and Y . Dai, “Uncertainty-aware Joint Salient Object and Camouflaged Object Detection,” arXiv.org, 2021. https://arxiv.org/abs/2104.02628 (accessed May 10, 2025)

    2021 arXiv

  62. [71]

    Segment, Magnify and Reiterate: Detecting Camouflaged Objects the Hard Way,

    Q. Jia, S. Yao, Y . Liu, X. Fan, R. Liu, and Z. Luo, “Segment, Magnify and Reiterate: Detecting Camouflaged Objects the Hard Way,” 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2022, doi: https://doi.org/10.1109/cvpr52688.2022.00467

    2022

  63. [72]

    Towards automatic polyp detection with a polyp appearance model,

    J. Bernal, J. S ´anchez, and F. Vilari ˜no, “Towards automatic polyp detection with a polyp appearance model,” Pattern Recognition, vol. 45, no. 9, pp. 3166–3182, Sep. 2012, doi: https://doi.org/10.1016/j.patcog.2012.03.002

    2012 doi

  64. [73]

    To- ward embedded detection of polyps in WCE images for early diag- nosis of colorectal cancer,

    J. Silva, A. Histace, O. Romain, X. Dray, and B. Granado, “To- ward embedded detection of polyps in WCE images for early diag- nosis of colorectal cancer,” International Journal of Computer Assisted Radiology and Surgery, vol. 9, no. 2, pp. 283–293, Sep. 2013, doi: https://doi...

    2013 doi

  65. [74]

    Kvasir-SEG: A Segmented Polyp Dataset,

    D. Jha et al., “Kvasir-SEG: A Segmented Polyp Dataset,” arXiv:1911.07069 [cs, eess], Nov. 2019, Available: https://arxiv.org/abs/1911.07069

    1911 arXiv

  66. [75]

    Automated Polyp Detection in Colonoscopy Videos Using Shape and Context Information,

    “Automated Polyp Detection in Colonoscopy Videos Using Shape and Context Information,” ieeexplore.ieee.org. https://ieeexplore.ieee.org/document/7294676

  67. [76]

    WM-DOV A maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians,

    J. Bernal, F. J. S ´anchez, G. Fern ´andez-Esparrach, D. Gil, C. Rodr ´ıguez, and F. Vilari ˜no, “WM-DOV A maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians,” Computer- ized Medical Imaging and Graphics, vol. 43, pp. 99–111, Jul....

    2015 doi

  68. [77]

    UNet++: A Nested U-Net Architecture for Medical Image Segmen- tation,

    Z. Zhou, M. M. Rahman Siddiquee, N. Tajbakhsh, and J. Liang, “UNet++: A Nested U-Net Architecture for Medical Image Segmen- tation,” Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, vol. 11045, pp. 3–11, 2018

    2018

  69. [78]

    Selective Feature Aggrega- tion Network with Area-Boundary Constraints for Polyp Segmentation,

    Y . Fang, C. Chen, Y . Yuan, and K. Tong, “Selective Feature Aggrega- tion Network with Area-Boundary Constraints for Polyp Segmentation,” Lecture notes in computer science, pp. 302–310

  70. [79]

    Adap- tive Context Selection for Polyp Segmentation,

    R. Zhang, G. Li, Z. Li, S. Cui, D. Qian, and Y . Yu, “Adap- tive Context Selection for Polyp Segmentation,” arXiv.org, 2023. https://arxiv.org/abs/2301.04799 (accessed May 10, 2025). Jianlin Sun received the BS degree, in 2024 from the Hefei University of technology, Hefei, ch...

    2023 arXiv

  71. [2021]

    https://arxiv.org/abs/2105.12555 (accessed May 09, 2025)

    2025 arXiv

  72. [2024]

    https://arxiv.org/abs/2409.01686 (accessed May 09, 2025). 12

    2025 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.