Writing study: WIR3D: Visually-Informed and Geometry-Aware 3D Shape Abstraction

ICCV overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← ICCV 2025 oral writing corpus

Richard Liu, Daniel Fu, Noah Tan, Itai Lang, Rana Hanocka · 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: Training, objectives & optimization (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

12 physical pages

Before references / appendix

≈ 8.04 page-equivalents

Main figures / tables

11 / 2

Attached appendix figures / tables

0 / 0

Attached appendix span

No boundary detected; not assumed zero

Before references / appendix: ≈ 67%References / intervening material: ≈ 33%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	≈ 1.26 pp.	10.5%
2. Related Work	2	≈ 1 pp.	8.33%
3. Method	3	≈ 2.09 pp.	17.42%
4. Experiments	5	≈ 3.02 pp.	25.17%
5. Conclusion	8	≈ 0.25 pp.	2.08%
References	9	≈ 3.96 pp.	33%

All detected figure and table locations

Main text

• Figure 1 · p. 1
• Figure 2 · p. 2
• Figure 3 · p. 3
• Figure 4 · p. 4
• Figure 5 · p. 5
• Figure 6 · p. 6
• Figure 7 · p. 6
• Table 1 · p. 6
• Figure 8 · p. 7
• Table 2 · p. 7
• Figure 9 · p. 7
• Figure 10 · p. 8
• Figure 11 · p. 8

Attached appendix

No recognized caption labels.

Argument and organization

1. 1 Introduction; 2 Related Work

   Uses a textured cylinder to distinguish sparse view-consistent abstraction from backprojected contours.

2. 3 Method

   Defines Bezier strokes, localized semantic/geometric losses and sequential geometry/texture optimization.

3. 4 Experiments; 5 Conclusion

   Combines perceptual/geometric comparisons, keypoint ablations and deformation applications before preprocessing limits.

Problem, claim and contribution

A simple cylinder counterexample explains why individually reasonable 2D contours fail to create a sparse 3D abstraction.

Figure 2 and Section 1, physical p.2 · Main paper

Argument and section order

Geometry is optimized first and frozen before new texture curves are introduced, making the separation of responsibilities explicit.

Section 3.3, p.5 · Main paper

Figures, captions and visual explanation

The two-stage overview draws different target renders and loss pathways; the stroke-count grid shows abstraction granularity without changing the target object.

Figure 3, p.3; Figure 6, p.6 · Main paper

Tables, comparisons and evidence

The comparison defines coverage separately from perceptual metrics and documents matched 40-curve budgets for 3Doodle, while acknowledging NEF cannot fix curve count.

Table 1 and Section 4.2, pp.6–7 · Main paper

Appendix and reproducibility

360-degree videos, interactive deformation, keypoint initialization, extra comparisons and ablation figures are routed to supplements; those were not inspected.

Sections 1,3.3,4.2–4.4, pp.2,5–8 · Main paper

Clarity and writing conventions

The paper explicitly warns that global CLIP similarity can miss substantial visual differences, motivating intermediate features and complementary evaluation.

Section 4.2, p.6 · Main paper

What to reuse—and what not to infer

Transferable lesson: Introduce abstraction with a minimal counterexample, then distinguish geometric coverage, perceptual similarity and human preference.

Boundary: Optional user keypoints and automatic keypoint quality influence results. Two in-the-wild examples do not prove arbitrary input robustness; preprocessing can take hours, and selected static views do not establish all-view consistency or interactive end-to-end speed.

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, 6

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 · 12 pages