Writing study: RAVEN: Erasing Invisible Watermarks via Novel View Synthesis

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Fahad Shamshad, Nils Lukas, Karthik Nandakumar · Official CVPR 2026 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: Empirical analysis, diagnosis & auditing (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

≈ 8.05 page-equivalents

Main figures / tables

5 / 6

Attached appendix figures / tables

0 / 0

Attached appendix span

Separate file recorded below; not in this PDF total

Before references / appendix: ≈ 80.5%References / intervening material: ≈ 19.5%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.

Source-bound correction: Table 3 on physical page 7 has a side caption whose label is a separate extracted line. Raster inspection confirms Table 3 and the source coordinate of that label; there are six main tables, not five.

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	≈ 1.28 pp.	12.8%
2. Related Work	2	≈ 0.62 pp.	6.2%
3. Threat Model	3	≈ 0.37 pp.	3.7%
4. Proposed Approach	3	≈ 2.67 pp.	26.7%
5. Experiments	6	≈ 2.56 pp.	25.6%
6. Conclusion	8	≈ 0.19 pp.	1.9%
References	9	≈ 1.95 pp.	19.5%

All detected figure and table locations

Main text

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

Attached appendix

No recognized caption labels.

Separate source files—excluded from canonical-PDF totals

A separate author version may repeat main text and references. Whole-file lengths and counts are not extra appendix material and are never added to the canonical PDF totals. Read the study's source scope for which sections received substantive reading versus layout inspection.

Additional source 1

Separate PDF (or ZIP containing it): whole-file layout of 5 physical pages; 4 detected figures and 1 tables. These may include overlapping main-text material.

Detected heading	PDF page	Estimated span
2. Quality–Detectability Trade-off Analysis	1	≈ 0.33 pp.
3. Quantitative Results on DiffusionDB	1	≈ 0.59 pp.
4. Additional Qualitative Results	2	≈ 0.27 pp.
5. Computational Efficiency	2	≈ 0.14 pp.
6. Defense Directions	2	≈ 2.24 pp.
References	5	≈ 0.95 pp.

• Figure 1 · p. 2
• Table 1 · p. 3
• Figure 2 · p. 3
• Figure 3 · p. 4
• Figure 4 · p. 4

Argument and organization

1. Problem and contribution

   Defines three competing goals—suppression, perceptual quality and compute—before proposing a view-synthesis interpretation.

2. Argument sequence

   States the threat model before partial inversion, viewpoint modulation, correspondence attention and color transfer.

3. Evidence and verification

   Distinguishes low detector true-positive rate from bit accuracy near chance, and reports image quality separately from removal effectiveness.

Problem, claim and contribution

Defines three competing goals—suppression, perceptual quality and compute—before proposing a view-synthesis interpretation.

Introduction and Table 1, p. 1 · Main paper · Additional source 2

Argument and section order

States the threat model before partial inversion, viewpoint modulation, correspondence attention and color transfer.

Sections 3–4, pp. 3–5 · Main paper · Additional source 2

Figures, captions and visual explanation

Figure 2 aligns full-image and zoomed comparison rows with red crop boxes; its caption explicitly warns that viewpoint shifts change some crop correspondence.

Figure 2, p. 7 · Main paper · Additional source 2

Tables, comparisons and evidence

Distinguishes low detector true-positive rate from bit accuracy near chance, and reports image quality separately from removal effectiveness.

Tables 2–4, pp. 6–7 · Main paper · Additional source 2

Appendix and reproducibility

The supplement adds quality–detectability curves and measured computational efficiency, supporting a trade-off claim rather than a single operating point.

Supplement Sections 2 and 5, pp. 1–2 · Main paper · Additional source 2

Clarity and writing conventions

Results use separate headings for removal performance and quality preservation, keeping two potentially conflicting claims legible.

Section 5.2, p. 7 · Main paper · Additional source 2

What to reuse—and what not to infer

Transferable lesson: Pair any attack or robustness score with the utility/distortion it costs, and describe alignment rules for qualitative crops.

Boundary: Introductory colored-dot ratings are qualitative summaries, not calibrated scores; preserve measured non-wins and the stated threat model.

Apply this lesson using the argument and evidence worksheet.

Inspection record and sources

Checked

2026-09-26

Reading status

Writing-focused manuscript scan + selected visuals inspected

Main text

Writing-oriented scan throughout Sections 1–6, PDF pp. 1–8; references pp. 9–10: abstract/framing, section sequence, figure/table captions, evaluation organization and conclusions; not an exhaustive technical verification.

Appendix / supplement

Writing-oriented supplement scan: Supplement Sections 1–6, pp. 1–5: baseline hyperparameters, quality/detectability curves, full DiffusionDB results, runtime and defense directions. Video/code assets were not executed or assessed.

Visually inspected PDF pages

7, 1

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 CVPR 2026 accepted-paper PDF · 10 pages
• Official CVF supplementary PDF · 5 pages