Writing study: Easy3D: A Simple Yet Effective Method for 3D Interactive Segmentation

ICCV overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← ICCV 2025 oral writing corpus

Andrea Simonelli, Norman Müller, Peter Kontschieder · Official ICCV 2025 oral record

Observed writing choices, not acceptance rules. This study examines the inspected manuscript version. It does not establish why the paper was accepted or that its presentation choices should be copied unchanged.

Figure, table and section ledger

Editorial contribution type: Architecture & representation (primary).

Layout extraction, not a required paper template. Counts are distinct caption labels; a multi-panel figure counts once. Section spans are estimated from heading positions, including floats and whitespace. Method and limitations.

Canonical PDF

10 physical pages

Before references / appendix

≈ 8.1 page-equivalents

Main figures / tables

5 / 4

Attached appendix figures / tables

0 / 0

Attached appendix span

No boundary detected; not assumed zero

Before references / appendix: ≈ 81%References / intervening material: ≈ 19%Estimated layout envelopes, not text/artwork area. Unresolved appendix boundaries remain within the undivided post-body material.

Extraction notes: No attached appendix boundary detected. Zero means no recognized caption, not independently verified absence.

Section-by-section lengths and source pages

The title/abstract/front matter precede the first numbered heading. A span ends at the next detected top-level heading, not at the last sentence; end-page whitespace is included. References and statements are listed separately. These are not measurements of text-only or artwork area.

Top-level heading	Starts on PDF page	Estimated span	Share of entire PDF
1. Introduction	1	≈ 0.97 pp.	9.7%
2. Related Works	2	≈ 0.56 pp.	5.6%
3. Method	2	≈ 2.99 pp.	29.9%
4. Experiments	5	≈ 2.86 pp.	28.6%
5. Qualitative Results	8	≈ 0.13 pp.	1.3%
6. Conclusions	8	≈ 0.21 pp.	2.1%
Acknowledgements	9	≈ 0.05 pp.	0.5%
References	9	≈ 1.9 pp.	19%

All detected figure and table locations

Main text

• Figure 1 · p. 1
• Figure 2 · p. 3
• Figure 3 · p. 4
• Table 1 · p. 6
• Table 2 · p. 7
• Table 3 · p. 7
• Table 4 · p. 7
• Figure 4 · p. 8
• Figure 5 · p. 8

Attached appendix

No recognized caption labels.

Argument and organization

1. 1 Introduction; 2 Related Works

   Positions sparse voxel encoding and implicit click fusion as complementary design choices.

2. 3 Method

   Defines the interaction loop, representation, positive/negative click handling and simulated-feedback training.

3. 4 Experiments; 5 Qualitative Results; 6 Conclusions

   Separates cross-domain interactive evaluation, single-click non-interactive comparison, fusion ablations and a VR demonstration.

Problem, claim and contribution

The desired output is one target instance selected by positive/negative user clicks, not automatic discovery of every object.

Section 3.1, physical pp.2–3 · Main paper

Argument and section order

Method detail follows the data path from point-to-voxel conversion to clicks/decoder/fusion and back to point masks, followed by the separate interaction simulator.

Sections 3.2–3.3, pp.3–5 · Main paper

Figures, captions and visual explanation

The architecture explicitly shows the voxel-to-point return mapping; qualitative rows place ground truth, click-specific masks and IoU at each interaction together.

Figure 2, p.3; Figure 4, p.8 · Main paper

Tables, comparisons and evidence

Evaluation uses ground-truth-aware simulated clicks selected from each method’s error regions, with up to ten clicks and matching training data for the main comparison.

Sections 3.3.1 and 4.3, pp.5–6; Table 1, p.6 · Main paper

Appendix and reproducibility

Exact decoder details, training settings and the derived Gaussian-Splatting dataset’s annotation transfer are deferred to supplementary material, not inspected.

Sections 3.2.4,3.3.3,4.1,4.7.2, pp.4–6,8 · Main paper

Clarity and writing conventions

The ablation discussion acknowledges explicit fusion can be preferable for many clicks in the training-like domain rather than claiming one design dominates unconditionally.

Sections 4.5 and 4.7.1, pp.6–7 · Main paper

What to reuse—and what not to infer

Transferable lesson: In interactive-method papers, make the feedback policy part of the experimental definition and show how each added interaction changes the output.

Boundary: Simulated oracle-error clicks are not a human usability study or a measurement of annotation time. Single-click comparisons with automatic methods have different input privileges. Gaussian-Splatting geometry is derived from ScanNet, not an independent new scene collection; the VR demonstration does not establish all latency or generalization claims.

Apply this lesson using the argument and evidence worksheet.

Inspection record and sources

Checked

2026-10-07

Reading status

Writing-focused manuscript scan + selected visuals inspected

Main text

Writing-oriented scan across the main manuscript, physical PDF pp.1–8, covering framing, section progression, method explanations, captions, comparison protocols and concluding scope; reference boundary checked. Selected full-page rasters were inspected. Not exhaustive line-by-line reading, proof verification or experimental reproduction.

Appendix / supplement

Main-paper appendix/supplement references were checked. A separate supplementary source is linked in the inventory but was not retrieved or inspected; its contents are not included in the reading or visual-page counts.

Visually inspected PDF pages

3, 8

This is a writing-focused scan of the manuscript and the stated supplementary scope: framing, organization, visual explanation, evidence presentation and wording. It is not a line-by-line proof audit, reproduction of results or exhaustive scientific review.

• Official CVF open-access ICCV 2025 accepted-paper PDF, retrieved 7 October 2026; CVF states identical to accepted version except watermark; not assumed identical to initial submission · 10 pages