- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes Integra-ICS, a training-free visual in-context segmentation framework built on frozen DINOv3 features. It aggregates foreground and background evidence through global top-k retrieval across support images, followed by candidate-guided query clustering and minimum-evidence filtering. Experiments on eight benchmarks show improved average segmentation performance in both 5-shot and 10-shot settings using a single hyperparameter configuration.
Critical strengths
-
The paper presents a clear motivation for aggregating multiple support correspondences rather than relying on a single nearest neighbor, addressing a plausible source of missed foreground regions.
-
Table 1 demonstrates strong average performance across eight diverse benchmarks, with improvements of 3.60 and 3.56 mIoU points over the strongest baseline on average in the 5-shot and 10-shot settings, respectively.
-
The framework requires no task-specific training and uses a single hyperparameter configuration across datasets. The runtime analysis also shows inference speed close to INSID3, suggesting that the additional aggregation introduces limited overhead under the evaluated setting.
Critical weaknesses
-
The related-work discussion overlooks relevant retrieval-based segmentation methods, such as “Shift the Lens: Environment-Aware Unsupervised Camouflaged Object Detection” (CVPR 2025) and “Beyond Single Images: Retrieval Self-Augmented Unsupervised Camouflaged Object Detection” (ICCV 2025). In particular, the latter constructs foreground/background prototype libraries and uses top-k nearest-neighbor retrieval and voting for segmentation, which is technically relevant to the proposed evidence aggregation. Although these methods address a different task setting, discussing their connections and differences would help clarify the novelty of the proposed approach.
-
The methodological novelty appears incremental. The framework combines global top-k retrieval, similarity-weighted aggregation, clustering, and filtering, but the manuscript does not sufficiently explain what distinguishes this combination from a straightforward nearest-neighbor label aggregation pipeline. The use of standard operations is not itself a weakness; the concern is that the distinctive technical insight and the necessity of the proposed combination remain insufficiently established.
-
The contribution over “INSID3: Training-Free In-Context Segmentation with DINOv3” (CVPR 2026) should be more clearly isolated through controlled comparisons. In particular, the core aggregation mechanism lacks simple baselines, such as uniform voting over the same top-k neighbors and per-shot top-k aggregation. Replacing only the correspondence module of INSID3 while keeping the remaining pipeline unchanged would also help determine how much of the improvement comes from the proposed evidence integration.
Questions for authors
-
What distinguishes the proposed retrieval and aggregation mechanism from prior retrieval-based segmentation approaches, particularly “Shift the Lens: Environment-Aware Unsupervised Camouflaged Object Detection” and “Beyond Single Images: Retrieval Self-Augmented Unsupervised Camouflaged Object Detection”? Beyond the different task setting, which technical contribution is specific to Integra-ICS?
-
Does similarity-weighted aggregation outperform uniform voting over the same top-k neighbors, and does global retrieval outperform per-shot top-k aggregation under otherwise identical conditions? Such comparisons would help establish whether the proposed aggregation provides benefits beyond considering additional neighbors.
-
Which components are inherited from INSID3, and which are newly introduced? Can the contribution of evidence integration be isolated by replacing only INSID3’s correspondence module while keeping the remaining pipeline unchanged?
-
What evidence distinguishes genuine cross-shot complementarity from redundant support or the general benefit of retrieving more tokens? In particular, does the CSCR analysis establish that contributions from different shots are necessary for the recovered predictions, rather than merely present among the retrieved neighbors?
3 comments
[W3] Controlled comparison with INSID3 and simple aggregation baselines
W3: The contribution over INSID3 should be isolated through controlled comparisons, including uniform voting over the same top-k neighbors and per-shot top-k aggregation, and by replacing only the correspondence module while keeping the remaining pipeline unchanged.
-
Answer: Thank you for this suggestion; we performed exactly this study. We disable ME and CMPE, replace only the correspondence module while keeping all other components fixed, and extend the evaluation from the five benchmarks of Table 3 to all eight.
Note: All variants share our base pipeline, and Variant A is not identical to
INSID3. It already includes shot-independent refinements: (i) features are L2-normalized both before and after positional debiasing, so that all similarities (matching, prototypes, and clustering) are cosine similarities in the same normalized debiased space, whereasINSID3uses original features for clustering and debiased features for matching; (ii) the forward-mask filtering of the officialINSID3implementation is removed, which we found to slightly hurt performance; and (iii) each shot's foreground prototype is normalized before averaging, so that every shot contributes equally regardless of its mask size. These refinements improve Variant A overINSID3† by+3.05mIoU (62.52vs.59.47) with the same clustering threshold (Appendix). We report them transparently; every gain attributed to our method is measured on top of this stronger baseline.Variant (5-shot) Correspondence Avg mIoU (8 benchmarks) Δ vs. A INSID3† (reference)Per-shot backward 1-NN 59.47 – A Per-shot backward 1-NN 62.52 – C Shot-wise top- (5 per shot) 62.43 −0.09 D Global top-, softmax 63.48 +0.96 B Global top-, uniform 63.64 +1.12 Full D + CMPE + ME 64.47 +1.95 Shot-wise top- (C) retrieves
25tokens per query token but yields no gain, whereas global top- (D, B) retrieves only5and improves by+0.96–+1.12. Since C and D differ only in whether candidates from different shots compete in a single ranking, the gain comes from integrating evidence across shots rather than from retrieving more tokens. Uniform voting (B) performs comparably, showing that the gain stems from cross-shot integration itself rather than a particular weighting scheme. We adopt softmax weighting so that each neighbor contributes according to its matching confidence, and will extend the comparison of weighting schemes in the revision. CMPE and ME add a further+0.99mIoU on top of D (64.47). We will add this study to the ablation section. -
Q1: What technically distinguishes Integra-ICS from EASE and RISE?
- Answer: Please see [W1]. Unlike EASE/RISE, which retrieve from external or mined libraries, Integra-ICS retrieves from the annotated FG/BG tokens of each episode's support shots, jointly across shots, to recover evidence that per-shot WTA discards.
-
Q2: Does similarity weighting beat uniform voting, and does global retrieval beat per-shot top-k?
- Answer: Please see the table in [W3] (ME/CMPE disabled, all else fixed). Global retrieval outperforms per-shot top- (
63.48/63.64vs.62.43), whereas similarity weighting and uniform voting are comparable (63.48vs.63.64). The gain thus comes from cross-shot integration, not the weighting rule; we will state this explicitly.
- Answer: Please see the table in [W3] (ME/CMPE disabled, all else fixed). Global retrieval outperforms per-shot top- (
-
Q3: Which components are inherited from INSID3, and can evidence integration be isolated?
- Answer: Inherited: the frozen DINOv3 backbone with positional debiasing, query-token clustering, and cluster-level scoring. New: global top- retrieval over a unified FG/BG pool (replacing per-shot 1-NN), CMPE, and ME. Replacing only the correspondence module improves the 8-benchmark average by
+0.96–+1.12mIoU, while shot-wise top- does not ([W3]). We will add a component breakdown table.
- Answer: Inherited: the frozen DINOv3 backbone with positional debiasing, query-token clustering, and cluster-level scoring. New: global top- retrieval over a unified FG/BG pool (replacing per-shot 1-NN), CMPE, and ME. Replacing only the correspondence module improves the 8-benchmark average by
-
Q4: Does CSCR show genuine cross-shot complementarity, rather than redundancy or simply more tokens?
- Answer: CSCR shows the presence of cross-shot evidence, not its complementarity. We therefore additionally measure the cosine similarity between FG tokens retrieved from different shots. On X-Ray, CSCR is highest (
74.39%), but this evidence is nearly identical (cosine0.952), and global retrieval has almost no effect (79.71→79.78). On the other datasets the evidence is less redundant and global retrieval helps on most of them. Retrieving more tokens alone is not enough: shot-wise top- (25tokens) does not beat 1-NN (62.43vs.62.52), whereas global top- with5tokens does (63.48). We will add the retrieved-cosine column to Table 2.
- Answer: CSCR shows the presence of cross-shot evidence, not its complementarity. We therefore additionally measure the cosine similarity between FG tokens retrieved from different shots. On X-Ray, CSCR is highest (
Thank you for the response. Most of my concerns have been addressed, and I have raised my score to 3.
[W1] Relation to retrieval-based segmentation methods (EASE, RISE)
Answer: We thank the reviewer for pointing out these relevant methods. We agree that EASE and RISE should have been cited and discussed, and we will add them to the related-work section in the revision. Although all three methods use foundation-model retrieval for foreground/background discrimination, EASE and RISE are developed for unsupervised camouflaged object detection, and neither is formulated around episodic visual-prompt inference or the loss of complementary evidence caused by winner-take-all (WTA) correspondence across annotated support shots. The main distinctions are summarized below.
EASE's self-retrieval also pools foreground/background prototypes under a coarse mask, but it builds a single image-level prototype per class and re-classifies each pixel by their contrast; CMPE instead computes a prototype for each query cluster from only its candidate tokens and uses it for cluster-level scoring against the support prototype.
We do not claim top-k retrieval itself as a new generic operation. Integra-ICS is an observation-driven contribution: we identify and quantify a specific bottleneck in multi-shot VP-ICS, i.e., foreground evidence distributed across support shots being discarded by WTA selection, and introduce our problem-specific mechanism to resolve it. Our controlled comparison (see [W3]) shows that what matters is where the retrieval is performed rather than how the retrieved labels are weighted: global retrieval over all shots improves over 1-NN with either softmax weighting (
63.48) or uniform voting (63.64), whereas shot-wise aggregation does not (62.43vs.62.52). We will revise the related-work and contribution descriptions accordingly.[W2] Distinctive insight beyond nearest-neighbor label aggregation
63.64vs.63.48). The distinctive insight therefore lies not in the voting rule but in two points that a straightforward nearest-neighbor label aggregation pipeline does not address. (1) Which neighbors compete. The bottleneck we identify is that per-shot WTA correspondence discards foreground evidence held by the other shots; ranking FG/BG tokens from all shots in a single list resolves it, whereas aggregating more neighbors within each shot does not (shot-wise top-k:−0.09despite five times more tokens). (2) How token evidence becomes a region decision. Token-level voting alone leaves foreground regions without direct support, so CMPE uses the candidate mask to guide cluster-level prototype extraction and reduce false negatives (Appendix Fig. 6), and because this expansion can introduce false positives, ME requires direct foreground support (Appendix Fig. 7). Each step addresses the error left by the previous one, which the 8-benchmark averages reflect: 1-NN62.52→ global top-k63.48→ with CMPE and ME64.47. We will make this progression and its rationale explicit in Section 3 of the revision.