Writing study: MeshSplatting: Differentiable Rendering with Opaque Meshes

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Jan Held, Sanghyun Son, Renaud Vandeghen, Daniel Rebain, Matheus Gadelha, Yi Zhou, Anthony Cioppa, Ming C. Lin, Marc Van Droogenbroeck, Andrea Tagliasacchi · 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: 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

≈ 8.05 page-equivalents

Main figures / tables

7 / 4

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.

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	≈ 0.59 pp.	5.9%
2. Related work	2	≈ 0.68 pp.	6.8%
3. Methodology	3	≈ 2.73 pp.	27.3%
4. Experiments	5	≈ 2.98 pp.	29.8%
5. Conclusions	8	≈ 0.26 pp.	2.6%
References	9	≈ 1.95 pp.	19.5%

All detected figure and table locations

Main text

• Figure 1 · p. 1
• Figure 2 · p. 3
• Figure 3 · p. 4
• Figure 4 · p. 5
• Table 1 · p. 6
• Table 2 · p. 6
• Figure 5 · p. 7
• Figure 6 · p. 7
• Table 3 · p. 8
• Table 4 · p. 8
• Figure 7 · 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 6 physical pages; 7 detected figures and 7 tables. These may include overlapping main-text material.

Located appendix only: starts on physical p.1; ≈ 5.94 page-equivalents, 7 figure labels and 7 table labels. This interval is distinct from the whole file.

Detected heading	PDF page	Estimated span
Supplementary Material	1	≈ 4.99 pp.
References	6	≈ 0.95 pp.

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

Argument and organization

1. Problem and contribution

   The contribution is not merely a higher rendering score: the introduction asks for meshes that remain connected, opaque and usable by conventional graphics tools. Those requirements become explicit comparison attributes rather than disappearing behind PSNR.

2. Argument sequence

   The method moves from rendering background to shared vertices, converting triangle soups into connected meshes, then opaque optimization. This separates representational changes from optimization stages before the experiments test both.

3. Evidence and verification

   Table 1 combines capability columns with reconstruction metrics, while Table 2 separately reports training time, rendering FPS at two resolutions and memory. The latter identifies MacBook M4 rendering, so deployment costs are not hidden inside a single quality number.

Problem, claim and contribution

The contribution is not merely a higher rendering score: the introduction asks for meshes that remain connected, opaque and usable by conventional graphics tools. Those requirements become explicit comparison attributes rather than disappearing behind PSNR.

Introduction; Table 1, PDF p.6 · Main paper · Additional source 2

Argument and section order

The method moves from rendering background to shared vertices, converting triangle soups into connected meshes, then opaque optimization. This separates representational changes from optimization stages before the experiments test both.

Sections 3.1–3.5; Section 4 · Main paper · Additional source 2

Figures, captions and visual explanation

Figure 2 contrasts independent triangles with shared-vertex updates, making the changed gradient interaction visible. Figure 3 then supplies the larger staged pipeline; using both prevents one crowded diagram from serving two explanatory jobs.

Figures 2–3, PDF pp.3–4 · Main paper · Additional source 2

Tables, comparisons and evidence

Table 1 combines capability columns with reconstruction metrics, while Table 2 separately reports training time, rendering FPS at two resolutions and memory. The latter identifies MacBook M4 rendering, so deployment costs are not hidden inside a single quality number.

Tables 1–2, PDF p.6 · Main paper · Additional source 2

Appendix and reproducibility

The supplementary sequence extends the main claims with implementation settings, per-scene scores, vertex-count and connectivity studies, additional geometry results and explicit limitations. This is a useful claim-indexed extension of the main table, not a second unexplained method.

Supplementary Sections 0.1–0.5, PDF pp.1–6 · Main paper · Additional source 2

Clarity and writing conventions

The paper distinguishes attractive renders from correct geometry: its regularization comparison explains that similar image scores can conceal undesirable surfaces. Preserve that distinction when captions discuss mesh quality.

Figure 7 and regularization analysis, main PDF p.8 · Main paper · Additional source 2

What to reuse—and what not to infer

Transferable lesson: Define the downstream artifact requirements first; then use one table for quality plus compatibility and another for real deployment costs.

Boundary: Opaque connected meshes are a design trade-off, not a guarantee of the best score in every scene or metric; keep non-winning rendering results and geometry failures visible.

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 main PDF pp.1–8 plus 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: all six supplementary PDF pages, with method details, per-scene results, connectivity/regularization ablations and limitations. Video/code assets were not executed or assessed.

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 CVPR 2026 accepted-paper PDF · 10 pages
• Official CVF supplementary PDF · 6 pages