Pith. sign in

REVIEW 4 major objections 5 minor 66 references

DiCoR: Decoupled Referent Disambiguation and Contour Recalibration for Efficient Referring Remote Sensing Image Segmentation

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read By splitting referent disambiguation from contour recalibration, DiCoR claims the best reported accuracy on three referring remote sensing segmentation benchmarks while staying in an efficient single-pass pipeline.

desk verdict Solid JFS-plus-modules paper with honest ablations; the 'best on three benchmarks' claim needs seeds and error bars before it can be trusted. read the letter →

arxiv 2608.12980 v1 pith:CH2J5MLB submitted 2026-08-13 cs.CV

classification cs.CV
keywords referringremotesensingimagesegmentationreferentdisambiguationcandidaterankingcontourrecalibrationresidualcorrectionefficientvision-languagefusionbenchmark
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 tries to close the accuracy-efficiency gap in referring remote sensing image segmentation (RRSIS), the task of outlining the object a natural-language expression points to in a satellite or aerial image. Current joint fusion models are fast but often pick the wrong referent or leave coarse boundaries, while decoupled prompt pipelines that call a large foundation segmenter are accurate but slow and memory-hungry. DiCoR claims both problems can be fixed inside a fast joint-fusion network by separating them into two supervised sub-tasks: a disambiguation-aware localization guidance module that ranks candidate regions under linguistic and geometric cues, and a lightweight contour recalibration module that predicts residual logit corrections only near boundaries. On RefSegRS, RRSIS-D, and RISBench the method reports the best mIoU and gIoU, and on RefSegRS it improves mIoU by 5.28 points and runs about 4.7 times faster than a foundation-model baseline. If the claim holds, accurate referring segmentation does not require a heavy external segmenter, which matters for real-time disaster response, monitoring, and mapping.

What carries the argument

The load-bearing object is the DLG-LCR decoupling inside a joint-fusion backbone. DLG (disambiguation-aware localization guidance) supplies the mechanism that resolves which object: it produces a candidate set from response peaks, ranks candidates with candidate-adaptive token reweighting (soft token weights $\omega_i$ from visual and geometric embeddings) plus a geometric consistency score, and feeds the best candidate's Gaussian-gated support back into $X_3$ as residual spatial recalibration. LCR (lightweight contour recalibration) supplies the mechanism that fixes where the boundary is: a compact encoder-decoder with mirrored skip connections receives the image and coarse sigmoid prediction $P$, emits logit residuals $\Delta Z$, and is trained with localized cross-entropy and Dice losses whose weight map $W(u)$ concentrates on contour-adjacent pixels. Together they convert two failure modes, ambiguity and contour imprecision, into separately supervised subproblems, which is what lets the model keep a single-pass, foundation-model-free inference pipeline.

What would settle it

Partition a benchmark's test set by whether the top-$K$ peaks of the response map $R$ overlap the ground-truth mask at all, and compare DiCoR's mIoU on the peak-misses subset versus the peak-hits subset; if accuracy on the miss subset does not fall toward the coarse-backbone baseline, DLG is recovering referents beyond its stated mechanism, and if it collapses, the response-estimator ceiling is confirmed.

Watch

Extended reading notes

Core claim

DiCoR's central claim is that decoupling referent resolution from mask refinement, rather than decoupling the entire pipeline from the model, is what makes segmentation accurate and cheap. The first module turns grounding into competition: a dense response estimator produces a response map $R=\Phi_{\mathrm{resp}}(X_3)$, a candidate generator keeps the top-$K$ peaks ($K=5$) under non-maximum suppression, and a candidate ranker scores each region by cosine similarity between its visual embedding and token-reweighted text plus a geometric term; the winning support is injected back into the fused feature by $X_3 \odot (1+\alpha C_{i^*})$. The second module, LCR, computes $\Delta Z=\Phi_{\mathrm{lcr}}(I\oplus P)$ and forms $\tilde{Z}=Z+\Delta Z$, learning only residual contour corrections under a region-aware loss that upweights boundary pixels. With distractor regions mined offline as hard negatives, DLG receives explicit ranking supervision, and LCR is pretrained on filtered, morphologically perturbed coarse masks from multiple checkpoints. The paper reports the best mIoU and gIoU on all three benchmarks, with a 5.28-point mIoU gain over MCD-Net on RefSegRS and a 21.25-point gain in Pr@0.9, attributing the high-threshold gains to LCR and the localization gains to DLG.

Load-bearing premise

The weakest link is the initial response map: if it never lights up near the true target, the later candidate ranking cannot rescue the answer, a ceiling the authors acknowledge in the Limitations section.

Editorial extensions

If this is right

  • If DiCoR's results hold, RRSIS can reach foundation-model-level mask quality without invoking a large segmenter at inference, making referring segmentation practical on resource-constrained platforms such as drones and onboard processors.
  • The 21-point gain at Pr@0.9 on RefSegRS implies contour recalibration is the main driver of high-IoU masks, so strict-quality downstream uses like precise building or vehicle delineation benefit most from LCR.
  • The decoupled training recipe, pretrain auxiliary modules on multi-checkpoint coarse outputs, filter by localization quality, then jointly adapt, offers a general template for adding task-specific modules to any trained joint-fusion pipeline.
  • DLG reformulates grounding as candidate ranking, so the same candidate-adaptive token reweighting and geometric scoring could be lifted into referring expression comprehension or detection with minimal architectural change.

Reading between the lines

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

  • The LCR residual-correction idea is not tied to RRSIS: any segmentation decoder that emits coarse logits could gain a plug-in boundary refiner, so the most direct testable extension is to apply it to natural-image referring segmentation or panoptic segmentation decoders.
  • DLG's reliance on SAM3-mined distractors suggests a training-efficiency bottleneck; one could test whether cheaper pseudo-distractors, such as other high-confidence mask proposals from the backbone itself, give the same ranking gains and make the approach less dependent on a very large external model.
  • Because the speed comparison is against current DPS implementations, the 4.7x figure may shrink if foundation segmenters are distilled or quantized; the accuracy comparison is more durable than the latency comparison.
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

4 major / 5 minor

Summary. The paper proposes DiCoR, a referring remote sensing image segmentation (RRSIS) method built on a joint fusion segmentation (JFS) backbone, with two auxiliary modules: a Disambiguation-aware Localization Guidance (DLG) module that ranks candidate regions from an intermediate response map, and a Lightweight Contour Recalibration (LCR) module that predicts residual corrections to coarse logits. DLG is trained with response supervision that includes SAM3-mined hard negatives and a ranking loss; LCR is trained in a decoupled stage on filtered coarse predictions under a contour-weighted loss. Experiments on RefSegRS, RRSIS-D, and RISBench report state-of-the-art mIoU and gIoU, with the largest gains on RefSegRS (mIoU 77.96 vs. 72.68 for MCD-Net), and the paper reports a favorable accuracy-efficiency trade-off versus DPS methods such as RSRefSeg-2. The code is released.

Significance. If the reported results hold, DiCoR is a meaningful contribution: it improves accuracy over strong JFS baselines while retaining a lightweight, single-pass inference pipeline, and it does so with two well-motivated, decoupled supervision mechanisms. The paper includes extensive ablations on RISBench, external benchmarks, and an efficiency analysis with an accuracy-efficiency index. The use of external SAM3 distractors and external benchmarks means the central accuracy claim is not circular. The main weakness is that the empirical ranking claim is supported by single-run numbers without variance or significance testing; on two of the three benchmarks the top-position margins are small (e.g., 0.22 mIoU on RRSIS-D), so the headline conclusion is not statistically grounded as presented.

major comments (4)
  1. [Tables II and III; Section IV-C] The central claim that DiCoR 'achieves the best segmentation accuracy across all three benchmarks' is not statistically established. On RISBench the reported mIoU margin over CroBIM-U is 0.68 points (70.30 vs. 69.62) and the gIoU margin over CSINet is 0.15 points (75.51 vs. 75.36); on RRSIS-D the mIoU margin over RS2-SAM 2 is 0.22 points (66.94 vs. 66.72) and the gIoU margin over BTDNet is 0.22 points (79.45 vs. 79.23). No standard deviations, number of seeds, or paired significance tests are reported for any table, including the ablations. If reruns shift by about half an mIoU point, DiCoR would no longer be the best method on two benchmarks. Please report results over multiple seeds with standard deviations and, ideally, paired significance tests or at least error bars, for both the main tables and the ablations.
  2. [Section III-E and Section IV-D (LCR training)] LCR is pretrained only on coarse predictions whose IoU lies in [0.5, 0.95), but at inference it is applied to all coarse predictions, including those with IoU below 0.5. This train/inference distribution mismatch means the module's behavior on severely mislocalized inputs is unmeasured; a coarse prediction that DLG fails to correct could be outside the LCR training distribution. The paper should either train LCR on the full distribution of coarse predictions or analyze performance conditioned on coarse IoU (e.g., report mIoU for test samples grouped by coarse-prediction IoU).
  3. [Limitations (unnumbered section before Conclusion)] The paper honestly states that when the response estimator fails to activate around the true referent, the candidate generator and ranker have limited ability to recover the correct target. This is a load-bearing ceiling for the DLG disambiguation claim. The paper should quantify how often this occurs in practice (e.g., the fraction of test samples where the true referent is not among the top-K peaks) and analyze whether DLG's reported gains are concentrated in the samples where the response map already contains the referent. Without such analysis, the disambiguation contribution is not verified on the cases where it is most needed.
  4. [Section III-C (SAM3 distractor mining)] The distractor supervision L_H is a key component of DLG; the ablation in Table VII shows it adds about 0.8 mIoU. However, the paper does not specify the SAM3 version, the input prompt used for proposal generation, the confidence threshold, the overlap threshold defining 'high confidence but low overlap', or the number of distractors retained per image. These details are needed for reproducibility and to understand the sensitivity of the method to the external mining process.
minor comments (5)
  1. [Abstract] The abstract states 'running 4.7% faster', while the body (Section IV-D and Table IV) reports a 4.7x speedup over RSRefSeg-2. The percentage sign appears to be a typo for the multiplication sign.
  2. [Table III] There is a missing space in the RSRefSeg-1 row: '48.5030.80' should be '48.50 30.80'. Also, 'Ours-' in the last row should be 'Ours'.
  3. [Table IV] The LA VT row appears to have a formatting issue: '28.6547.40 47.40' should probably be '28.65 47.40 47.40'.
  4. [Figure 1 caption] The asterisk for CroBIM* is explained in the caption, but it is unclear what 'reimplementation' means here; please clarify whether the result was obtained by re-running the released code or by re-implementing the method.
  5. [Section IV-D, Eq. (20)] The accuracy-efficiency index AEI is a new composite metric with arbitrary normalization by the best FPS and lowest GFLOPs in the comparison set. Please justify this choice and, if possible, also report raw FPS and GFLOPs (which are already given) so readers can assess the trade-off without a single index.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: DiCoR's central claims are evaluated on external benchmarks, and the LCR self-distillation is training-data construction rather than a prediction derived from its own conclusion.

full rationale

The paper's central claim is an empirical ranking: DiCoR achieves the best mIoU/gIoU on RefSegRS, RISBench, and RRSIS-D against a broad set of external baselines. These evaluations are standard benchmark comparisons with publicly available implementations and are not constructed from DiCoR's own outputs, so no fitted parameter is renamed as a prediction. DLG's candidate ranking is supervised by SAM3-mined distractors, an external frozen model, and the ranking loss Lrank only encourages the model to prefer the ground-truth candidate among top-K responses; it does not predetermine the reported test accuracy. LCR is trained on coarse predictions collected from the same model's intermediate checkpoints and filtered by localization quality, which is a self-distillation-style training-data construction. This is self-referential in data sourcing but not circular: the final masks on held-out test examples are not defined as the coarse predictions, and the residual correction network is evaluated against ground truth, not against its own training targets. The Limitations section honestly states that DLG cannot recover a referent when the response estimator fails to activate, which is a performance ceiling, not a circular argument. No load-bearing step reduces, by the paper's own equations or by self-citation, to its own inputs. Therefore the appropriate finding is no significant circularity, score 0.

Assumptions & free parameters 10 free parameters · 4 assumptions · 0 invented entities

The method is an empirical deep learning system, so the ledger is mostly hyperparameters and domain assumptions. K, sigma_c, alpha, and the lambda weights are hand-chosen or tuned on validation data. The key assumptions are that the response map peaks near the referent, that SAM3 proposals form useful hard negatives, that LCR's training filter generalizes to all test-time coarse masks, and that a ground-truth candidate index exists among the top-K peaks for the ranking loss.

free parameters (10)
  • K = 5
    Number of candidate peaks retained by NMS-based peak selection; the ablation shows K=5 is best.
  • sigma_c = 3
    Gaussian support scale for candidate aggregation in Eq. 3, chosen for small remote sensing objects.
  • alpha = 0.5
    Residual spatial guidance strength in Eq. 8.
  • lambda_g = 0.5
    Weight for geometric consistency in the candidate score, Eq. 6.
  • lambda_resp = 0.9
    Weight for response supervision in the DLG loss, Eq. 11.
  • lambda_rank = 1.1
    Weight for the ranking loss in the DLG loss, Eq. 11.
  • lambda_ce = 1.0
    Weight for region-aware cross-entropy in the LCR loss, Eq. 14.
  • lambda_dice = 1.0
    Weight for region-aware Dice in the LCR loss, Eq. 14.
  • LCR IoU filtering threshold = [0.5, 0.95)
    Range of coarse prediction IoU retained for LCR pretraining; creates a train/test distribution mismatch because inference is not filtered.
  • checkpoint epochs for DLG/LCR pretraining = 10, 20, 25, 30, 39
    Intermediate checkpoints sampled to diversify coarse priors for the auxiliary modules.
assumptions (4)
  • domain assumption The true referent is expected to appear as a peak in the response map R.
    The DLG candidate generator only sees top-K peaks of R; the paper's Limitations section concedes failure when R does not activate on the target.
  • domain assumption SAM3-mined object proposals are reliable hard-negative distractors for response supervision.
    Hard instance background H is built from offline SAM3 proposals with high confidence and low overlap with ground truth; the exact confidence and overlap thresholds are not specified.
  • domain assumption LCR's training filter (coarse IoU in [0.5,0.95)) transfers to all test-time coarse masks.
    At inference LCR is applied to every prediction, including coarse masks with IoU below 0.5, but the paper only trains on filtered samples.
  • domain assumption A ground-truth candidate index y is available among the top-K candidates for the ranking loss.
    Equation 10 assumes y is defined; the paper does not specify how samples where the referent is not among the top-K peaks are handled.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DiCoR: Decoupled Referent Disambiguation and Contour Recalibration for Efficient Referring Remote Sensing Image Segmentation." pith.science (2026). https://pith.science/paper/CH2J5MLB

@misc{pith2026260812980,
  author       = {Pith},
  title        = {Pith review of: DiCoR: Decoupled Referent Disambiguation and Contour Recalibration for Efficient Referring Remote Sensing Image Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CH2J5MLB}},
  note         = {Machine review of arXiv:2608.12980}
}
read the original abstract

Referring remote sensing image segmentation (RRSIS) aims to delineate targets specified by natural language expressions in remote sensing imagery. Existing methods mainly follow joint fusion segmentation (JFS) or decoupled prompt segmentation (DPS). JFS is efficient but often suffers from limited accuracy because referent localization and mask delineation are optimized under a unified objective, whereas DPS separates localization from mask generation using spatial prompts and foundation segmenters at the cost of higher memory consumption and inference latency. To bridge this gap, we propose DiCoR, a decoupled referent disambiguation and contour recalibration framework built on an efficient JFS pipeline. DiCoR addresses two key challenges: distinguishing the correct referent from ambiguous candidates and refining coarse masks after localization. A disambiguation-aware localization guidance strategy ranks salient candidate regions with adaptive linguistic cues and injects the resulting localization prior into fused features. A lightweight contour recalibration module further predicts residual corrections to coarse logits under localized contour supervision, improving mask quality with limited computational overhead. Experiments on RefSegRS, RRSIS-D, and RISBench show that DiCoR achieves the best segmentation accuracy across all three benchmarks. On RefSegRS, it improves mIoU and gIoU by 5.28% and 2.87% over a competitive JFS method while running 4.7% faster than a representative DPS method, demonstrating a favorable accuracy-efficiency trade-off. Code is available at https://github.com/zyGao1126/DiCoR.

Figures

Figures reproduced from arXiv: 2608.12980 by the authors.

Figure 1
Figure 1. Accuracy–efficiency trade-off on the RefSegRS bench [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Comparison of conventional JFS pipeline and DiCoR. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Overview of the proposed DiCoR framework, including (a) the overall pipeline built upon a JFS backbone, (b) [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Structure of the MSA module [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 6
Figure 6. Figure 6: Architecture of the proposed DLG module. DLG pro [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 5
Figure 5. Figure 5: Failure cases of the JFS backbone. From left to right [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 7
Figure 7. Figure 7: Structure of the LCR module. Let Z denote the coarse segmentation logits produced by the decoder and P = σ(Z) denote the corresponding mask prediction through sigmoid activation, the LCR module takes the input image I and P as guidance to estimate a residual correction…
Figure 8
Figure 8. Figure 8: Qualitative results of the proposed DiCoR on the RefSegRS dataset. [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: Qualitative results of the proposed DiCoR on the RISBench dataset. [PITH_FULL_IMAGE:figures/full_fig_p010_9.png]
Figure 10
Figure 10. Figure 10: Qualitative results of the proposed DiCoR on the RRSIS-D dataset. [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 12
Figure 12. Figure 12: Ablation of LCR training data construction. Single [PITH_FULL_IMAGE:figures/full_fig_p013_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

66 extracted references · 41 canonical work pages

  1. [1]

    Rrsis: Referring remote sensing image segmentation,

    Z. Yuan, L. Mou, Y . Hua, and X. X. Zhu, “Rrsis: Referring remote sensing image segmentation,”IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1–12, 2024

  2. [2]

    Rotated multi-scale interaction network for referring remote sensing image seg- mentation,

    S. Liu, Y . Ma, X. Zhang, H. Wang, J. Ji, X. Sun, and R. Ji, “Rotated multi-scale interaction network for referring remote sensing image seg- mentation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 26 658–26 668

  3. [3]

    Cross-modal bidirectional interaction model for referring remote sensing image segmentation,

    Z. Dong, Y . Sun, T. Liu, W. Zuo, and Y . Gu, “Cross-modal bidirectional interaction model for referring remote sensing image segmentation,” arXiv preprint arXiv:2410.08613, 2024

  4. [4]

    Exploring fine-grained image-text alignment for referring remote sensing image segmentation,

    S. Lei, X. Xiao, T. Zhang, H.-C. Li, Z. Shi, and Q. Zhu, “Exploring fine-grained image-text alignment for referring remote sensing image segmentation,”IEEE Transactions on Geoscience and Remote Sensing, vol. 63, pp. 1–11, 2025

  5. [5]

    A large-scale referring remote sensing image segmentation dataset and benchmark,

    Z. Yang, H. Yao, L. Tian, X. Zhao, Q. Li, and Q. Wang, “A large-scale referring remote sensing image segmentation dataset and benchmark,” arXiv preprint arXiv:2506.03583, 2025

  6. [6]

    Geopix: A multimodal large language model for pixel-level image understanding in remote sensing,

    R. Ou, Y . Hu, F. Zhang, J. Chen, and Y . Liu, “Geopix: A multimodal large language model for pixel-level image understanding in remote sensing,”IEEE Geoscience and Remote Sensing Magazine, 2025

  7. [7]

    Rrsecs: Referring remote sensing expression com- prehension and segmentation,

    X. Lu, L. Sun, L. Li, L. Jiao, Y . Yang, Z. Huang, J. Chai, X. Liu, F. Liu, W. Maet al., “Rrsecs: Referring remote sensing expression com- prehension and segmentation,”IEEE Geoscience and Remote Sensing Magazine, 2025

  8. [8]

    Enabling training- free text-based remote sensing segmentation,

    J. Sosa, D. Rukhovich, A. Kacem, and D. Aouada, “Enabling training- free text-based remote sensing segmentation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2026, pp. 8041–8052

Show all 66 references
  1. [9]

    Real-time semantic segmen- tation: A brief survey and comparative study in remote sensing,

    C. Broni-Bediako, J. Xia, and N. Yokoya, “Real-time semantic segmen- tation: A brief survey and comparative study in remote sensing,”IEEE Geoscience and Remote Sensing Magazine, vol. 11, no. 4, pp. 94–124, 2023

  2. [10]

    Light-weight semantic segmentation network for uav remote sensing images,

    S. Liu, J. Cheng, L. Liang, H. Bai, and W. Dang, “Light-weight semantic segmentation network for uav remote sensing images,”IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 8287–8296, 2021

  3. [11]

    Rethinking the implicit optimization paradigm with dual alignments for referring remote sensing image segmentation,

    Y . Pan, R. Sun, Y . Wang, T. Zhang, and Y . Zhang, “Rethinking the implicit optimization paradigm with dual alignments for referring remote sensing image segmentation,” inProceedings of the 32nd ACM International Conference on Multimedia, 2024, pp. 2031–2040

  4. [12]

    Cadformer: Fine-grained cross- modal alignment and decoding transformer for referring remote sensing image segmentation,

    M. Liu, X. Jiang, and X. Zhang, “Cadformer: Fine-grained cross- modal alignment and decoding transformer for referring remote sensing image segmentation,”IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025

  5. [13]

    Lscf: Long-term semantic-guidance convformer for referring remote sensing image segmentation,

    Q. Ma, L. Li, X. Lu, L. Jiao, F. Liu, W. Ma, X. Liu, and L. Sun, “Lscf: Long-term semantic-guidance convformer for referring remote sensing image segmentation,”IEEE Transactions on Geoscience and Remote Sensing, 2025

  6. [14]

    Scale-wise bidirectional align- ment network for referring remote sensing image segmentation,

    K. Li, G. V osselman, and M. Y . Yang, “Scale-wise bidirectional align- ment network for referring remote sensing image segmentation,”ISPRS Journal of Photogrammetry and Remote Sensing, vol. 226, pp. 350–363, 2025

  7. [15]

    Referring remote sensing image segmentation via bidirectional alignment guided joint prediction,

    T. Zhang, Z. Wen, B. Kong, K. Liu, Y . Zhang, P. Zhuang, and J. Li, “Referring remote sensing image segmentation via bidirectional alignment guided joint prediction,”arXiv preprint arXiv:2502.08486, 2025

  8. [16]

    Referring remote sensing image segmen- tation with cross-view semantics interaction network,

    J. Yang, L. Zhang, and H. Lu, “Referring remote sensing image segmen- tation with cross-view semantics interaction network,”arXiv preprint arXiv:2508.01331, 2025

  9. [17]

    Crobim-u: Uncertainty- driven referring remote sensing image segmentation,

    Y . Sun, Z. Dong, H. Jiang, Y . Gu, and T. Liu, “Crobim-u: Uncertainty- driven referring remote sensing image segmentation,”IEEE Transactions on Geoscience and Remote Sensing, 2026

  10. [18]

    A multiscale vision–text collaborative dual encoder for referring rs image segmentation,

    J. Zhang, L. Li, L. Jiao, X. Liu, F. Liu, W. Ma, and S. Yang, “A multiscale vision–text collaborative dual encoder for referring rs image segmentation,”IEEE Transactions on Geoscience and Remote Sensing, vol. 64, pp. 1–15, 2026

  11. [19]

    Rsrefseg: Referring remote sensing image segmentation with foundation models,

    K. Chen, J. Zhang, C. Liu, Z. Zou, and Z. Shi, “Rsrefseg: Referring remote sensing image segmentation with foundation models,” inIGARSS 2025-2025 IEEE International Geoscience and Remote Sensing Sympo- sium. IEEE, 2025, pp. 1070–1074

  12. [20]

    Rsrefseg 2: decoupling referring remote sensing image segmentation with foundation models,

    K. Chen, C. Liu, B. Chen, J. Zhang, Z. Zou, and Z. Shi, “Rsrefseg 2: decoupling referring remote sensing image segmentation with foundation models,”arXiv preprint arXiv:2507.06231, 2025

  13. [21]

    Segearth-r1: Geospatial pixel reasoning via large language model,

    K. Li, Z. Xin, L. Pang, C. Pang, Y . Deng, J. Yao, G. Xia, D. Meng, Z. Wang, and X. Cao, “Segearth-r1: Geospatial pixel reasoning via large language model,”arXiv preprint arXiv:2504.09644, 2025

  14. [22]

    Segearth-r2: Towards comprehensive language-guided seg- mentation for remote sensing images,

    Z. Xin, K. Li, L. Chen, W. Li, Y . Xiao, H. Qiao, W. Zhang, D. Meng, and X. Cao, “Segearth-r2: Towards comprehensive language-guided seg- mentation for remote sensing images,”arXiv preprint arXiv:2512.20013, 2025

  15. [23]

    Customized sam 2 for referring remote sensing image segmentation,

    F. Rong, M. Lan, Q. Zhang, and L. Zhang, “Customized sam 2 for referring remote sensing image segmentation,”arXiv e-prints, pp. arXiv– 2503, 2025

  16. [24]

    Sam 3: Segment anything with concepts,

    N. Carion, L. Gustafson, Y .-T. Hu, S. Debnath, R. Hu, D. Suris, C. Ryali, K. V . Alwala, H. Khedr, A. Huanget al., “Sam 3: Segment anything with concepts,”arXiv preprint arXiv:2511.16719, 2025

  17. [25]

    Segmentation from natural language expressions,

    R. Hu, M. Rohrbach, and T. Darrell, “Segmentation from natural language expressions,” inEuropean Conference on Computer Vision. Springer, 2016, pp. 108–124

  18. [26]

    Referring image segmentation via recurrent refinement networks,

    R. Li, K. Li, Y .-C. Kuo, M. Shu, X. Qi, X. Shen, and J. Jia, “Referring image segmentation via recurrent refinement networks,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 5745–5753

  19. [27]

    Dynamic multimodal instance segmentation guided by natural language queries,

    E. Margffoy-Tuay, J. C. Perez, E. Botero, and P. Arbelaez, “Dynamic multimodal instance segmentation guided by natural language queries,” inEuropean Conference on Computer Vision. Springer, 2018, pp. 630– 645

  20. [28]

    Cross-modal self-attention network for referring image segmentation,

    L. Ye, M. Rochan, Z. Liu, and Y . Wang, “Cross-modal self-attention network for referring image segmentation,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 10 502–10 511

  21. [29]

    Bi-directional relationship inferring network for referring image segmentation,

    Z. Hu, G. Feng, J. Sun, L. Zhang, and H. Lu, “Bi-directional relationship inferring network for referring image segmentation,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 4424–4433

  22. [30]

    Multi- task collaborative network for joint referring expression comprehension and segmentation,

    G. Luo, Y . Zhou, X. Sun, L. Cao, C. Wu, C. Deng, and R. Ji, “Multi- task collaborative network for joint referring expression comprehension and segmentation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 10 034–10 043

  23. [31]

    Encoder fusion network with co- attention embedding for referring image segmentation,

    G. Feng, Z. Hu, L. Zhang, and H. Lu, “Encoder fusion network with co- attention embedding for referring image segmentation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2021, pp. 15 506–15 515

  24. [32]

    Vision-language transformer and query generation for referring segmentation,

    H. Ding, C. Liu, S. Wang, and X. Jiang, “Vision-language transformer and query generation for referring segmentation,” inProceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 16 321–16 330

  25. [33]

    Restr: Convolution- free referring image segmentation using transformers,

    N. Kim, D. Kim, C. Lan, W. Zeng, and S. Kwak, “Restr: Convolution- free referring image segmentation using transformers,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 18 145–18 154

  26. [34]

    Lavt: Language-aware vision transformer for referring image segmentation,

    Z. Yang, J. Wang, Y . Tang, K. Chen, H. Zhao, and P. H. Torr, “Lavt: Language-aware vision transformer for referring image segmentation,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 18 155–18 165

  27. [35]

    Slvit: Scale-wise language-guided vision transformer for referring image segmentation

    S. Ouyang, H. Wang, S. Xie, Z. Niu, R. Tong, Y .-W. Chen, and L. Lin, “Slvit: Scale-wise language-guided vision transformer for referring image segmentation.” inIJCAI, vol. 8, 2023

  28. [36]

    Caris: Context-aware referring image segmentation,

    S.-A. Liu, Y . Zhang, Z. Qiu, H. Xie, Y . Zhang, and T. Yao, “Caris: Context-aware referring image segmentation,” inProceedings of the 31st ACM International Conference on Multimedia, 2023, pp. 779–788. 15

  29. [37]

    Cris: Clip-driven referring image segmentation,

    Z. Wang, Y . Lu, Q. Li, X. Tao, Y . Guo, M. Gong, and T. Liu, “Cris: Clip-driven referring image segmentation,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 11 686–11 695

  30. [38]

    Extending clip’s image-text alignment to referring image segmentation,

    S. Kim, M. Kang, D. Kim, J. Park, and S. Kwak, “Extending clip’s image-text alignment to referring image segmentation,” inProceedings of the 2024 Conference of the North American Chapter of the Asso- ciation for Computational Linguistics: Human Language Technologies (Volume 1:...

  31. [39]

    Prompt-driven referring image segmentation with instance contrasting,

    C. Shang, Z. Song, H. Qiu, L. Wang, F. Meng, and H. Li, “Prompt-driven referring image segmentation with instance contrasting,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2024, pp. 4124–4134

  32. [40]

    LISA: Reasoning segmentation via large language model,

    X. Lai, Z. Tian, Y . Chen, Y . Li, Y . Yuan, S. Liu, and J. Jia, “LISA: Reasoning segmentation via large language model,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 9579–9589

  33. [41]

    Bridging vision and language encoders: Parameter-efficient tuning for referring image segmentation,

    Z. Xu, Z. Chen, Y . Zhang, Y . Song, X. Wan, and G. Li, “Bridging vision and language encoders: Parameter-efficient tuning for referring image segmentation,” inProceedings of the IEEE/CVF international conference on computer vision, 2023, pp. 17 503–17 512

  34. [42]

    Zero-shot referring image segmentation with global-local context features,

    S. Yu, P. H. Seo, and J. Son, “Zero-shot referring image segmentation with global-local context features,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 19 456–19 465

  35. [43]

    Evf-sam: Early vision-language fusion for text- prompted segment anything model,

    Y . Zhang, T. Cheng, L. Zhu, R. Hu, L. Liu, H. Liu, L. Ran, X. Chen, W. Liu, and X. Wang, “Evf-sam: Early vision-language fusion for text- prompted segment anything model,”arXiv preprint arXiv:2406.20076, 2024

  36. [44]

    Feature design for bridging SAM and CLIP toward referring image segmentation,

    K. Ito, “Feature design for bridging SAM and CLIP toward referring image segmentation,” inProceedings of the Winter Conference on Applications of Computer Vision, 2025, pp. 8357–8367

  37. [45]

    Gated-SCNN: Gated shape CNNs for semantic segmentation,

    T. Takikawa, D. Acuna, V . Jampani, and S. Fidler, “Gated-SCNN: Gated shape CNNs for semantic segmentation,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 5229–5238

  38. [46]

    SegFix: Model-agnostic boundary refinement for segmentation,

    Y . Yuan, J. Xie, X. Chen, and J. Wang, “SegFix: Model-agnostic boundary refinement for segmentation,” inEuropean Conference on Computer Vision. Springer, 2020, pp. 489–506

  39. [47]

    Boundary IoU: Improving object-centric image segmentation evaluation,

    B. Cheng, R. Girshick, P. Dollar, A. C. Berg, and A. Kirillov, “Boundary IoU: Improving object-centric image segmentation evaluation,” inPro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 15 334–15 342

  40. [48]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clarket al., “Learning transferable visual models from natural language supervision,” inInternational conference on machine learning. PmLR, 2021, pp. 8748–8763

  41. [49]

    Grounded language-image pre-training,

    L. H. Li, P. Zhang, H. Zhang, J. Yang, C. Li, Y . Zhong, L. Wang, L. Yuan, L. Zhang, J.-N. Hwang, K.-W. Chang, and J. Gao, “Grounded language-image pre-training,” inProceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, 2022, pp. 10 965– 10 975

  42. [50]

    Grounding DINO: Marrying DINO with grounded pre-training for open-set object detection,

    S. Liu, Z. Zeng, T. Ren, F. Li, H. Zhang, J. Yang, Q. Jiang, C. Li, J. Yang, H. Su, J. Zhu, and L. Zhang, “Grounding DINO: Marrying DINO with grounded pre-training for open-set object detection,” in European Conference on Computer Vision. Springer, 2024, pp. 38– 55

  43. [51]

    RemoteCLIP: A vision language foundation model for remote sensing,

    F. Liu, D. Chen, Z. Guan, X. Zhou, J. Zhu, Q. Ye, L. Fu, and J. Zhou, “RemoteCLIP: A vision language foundation model for remote sensing,” IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1– 16, 2024

  44. [52]

    Segment anything,

    A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y . Loet al., “Segment anything,” inProceedings of the IEEE/CVF international conference on computer vision, 2023, pp. 4015–4026

  45. [53]

    Segment everything everywhere all at once,

    X. Zou, J. Yang, H. Zhang, F. Li, L. Li, J. Wang, L. Wang, J. Gao, and Y . J. Lee, “Segment everything everywhere all at once,” inAdvances in Neural Information Processing Systems, vol. 36, 2023, pp. 19 769– 19 782

  46. [54]

    Segment anything in high quality,

    L. Ke, M. Ye, M. Danelljan, Y . Liu, Y .-W. Tai, C.-K. Tang, and F. Yu, “Segment anything in high quality,” inAdvances in Neural Information Processing Systems, vol. 36, 2023, pp. 29 914–29 934

  47. [55]

    SAM 2: Segment anything in images and videos,

    N. Ravi, V . Gabeur, Y .-T. Hu, R. Hu, C. Ryali, T. Ma, H. Khedr, R. Radle, C. Rolland, L. Gustafson, E. Mintun, J. Pan, K. V . Alwala, N. Carion, C.-Y . Wu, R. Girshick, P. Dollar, and C. Feichtenhofer, “SAM 2: Segment anything in images and videos,” inInternational Conferenc...

  48. [56]

    Semantic localization guiding seg- ment anything model for reference remote sensing image segmentation,

    S. Li, S. Wang, Z. Sun, and J. Xiao, “Semantic localization guiding seg- ment anything model for reference remote sensing image segmentation,” arXiv preprint arXiv:2506.10503, 2025

  49. [57]

    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

  50. [58]

    Bert: Pre-training of deep bidirectional transformers for language understanding,

    J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “Bert: Pre-training of deep bidirectional transformers for language understanding,” inPro- ceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technolog...

  51. [59]

    Linguistic structure guided context modeling for referring image segmentation,

    T. Hui, S. Liu, S. Huang, G. Li, S. Yu, F. Zhang, and J. Han, “Linguistic structure guided context modeling for referring image segmentation,” in European conference on computer vision. Springer, 2020, pp. 59–75

  52. [60]

    Cross-modal pro- gressive comprehension for referring segmentation,

    S. Liu, T. Hui, S. Huang, Y . Wei, B. Li, and G. Li, “Cross-modal pro- gressive comprehension for referring segmentation,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 9, pp. 4761– 4775, 2021

  53. [61]

    Beyond one-to-one: Rethinking the referring image segmentation,

    Y . Hu, Q. Wang, W. Shao, E. Xie, Z. Li, J. Han, and P. Luo, “Beyond one-to-one: Rethinking the referring image segmentation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 4067–4077

  54. [62]

    Cross-aware early fusion with stage- divided vision and language transformer encoders for referring image segmentation,

    Y . Cho, H. Yu, and S.-J. Kang, “Cross-aware early fusion with stage- divided vision and language transformer encoders for referring image segmentation,”IEEE Transactions on Multimedia, vol. 26, pp. 5823– 5833, 2023

  55. [63]

    Refer- ring image segmentation via cross-modal progressive comprehension,

    S. Huang, T. Hui, S. Liu, G. Li, Y . Wei, J. Han, L. Liu, and B. Li, “Refer- ring image segmentation via cross-modal progressive comprehension,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 10 488–10 497

  56. [64]

    Toward robust referring image segmentation,

    J. Wu, X. Li, X. Li, H. Ding, Y . Tong, and D. Tao, “Toward robust referring image segmentation,”IEEE Transactions on Image Processing, vol. 33, pp. 1782–1794, 2024

  57. [65]

    MnasNet: Platform-aware neural architecture search for mobile,

    M. Tan, B. Chen, R. Pang, V . Vasudevan, M. Sandler, A. Howard, and Q. V . Le, “MnasNet: Platform-aware neural architecture search for mobile,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 2815–2823

  58. [66]

    EfficientViM: Efficient vision mamba with hidden state mixer based state space duality,

    S. Lee, J. Choi, and H. J. Kim, “EfficientViM: Efficient vision mamba with hidden state mixer based state space duality,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025, pp. 14 923–14 933

Pith tools

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