Writing study: SmokeSVD: Smoke Reconstruction from A Single View via Progressive Novel View Synthesis and Refinement with Diffusion Models
← CVPR 2026 oral writing corpus
Chen Li, Shanshan Dong, Sheng Qiu, Jianmin Han, Yibo Zhao, Zan Gao, Taku Komura, Kemeng Huang · Official CVPR 2026 oral record
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
- 11 physical pages
- Before references / appendix
- ≈ 8.15 page-equivalents
- Main figures / tables
- 16 / 6
- Attached appendix figures / tables
- 0 / 0
- Attached appendix span
- Separate file recorded below; not in this PDF total
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 | ≈ 0.72 pp. | 6.55% |
| 2. Related Work | 2 | ≈ 0.5 pp. | 4.55% |
| 3. Method | 3 | ≈ 2.76 pp. | 25.09% |
| 4. Evaluations and Ablation Study | 5 | ≈ 3 pp. | 27.27% |
| 5. Conclusion and Future Work | 8 | ≈ 0.12 pp. | 1.09% |
| Acknowledgements | 8 | ≈ 0.23 pp. | 2.09% |
| References | 9 | ≈ 2.85 pp. | 25.91% |
All detected figure and table locations
Main text
- Figure 1 · p. 1
- Figure 3 · p. 3
- Figure 2 · p. 4
- Figure 4 · p. 5
- Figure 5 · p. 6
- Figure 6 · p. 6
- Table 1 · p. 6
- Table 2 · p. 6
- Table 3 · p. 6
- Figure 7 · p. 6
- Figure 8 · p. 7
- Figure 9 · p. 7
- Table 4 · p. 7
- Figure 10 · p. 7
- Figure 11 · p. 7
- Table 5 · p. 7
- Figure 14 · p. 7
- Figure 12 · p. 8
- Figure 13 · p. 8
- Figure 15 · p. 8
- Table 6 · p. 8
- Figure 16 · 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 10 physical pages; 17 detected figures and 4 tables. These may include overlapping main-text material.
Located appendix only: starts on physical p.1; ≈ 9.93 page-equivalents, 17 figure labels and 4 table labels. This interval is distinct from the whole file.
| Detected heading | PDF page | Estimated span |
|---|---|---|
| Supplementary Material | 1 | ≈ 0.02 pp. |
| A. Overview | 1 | ≈ 0.06 pp. |
| B. Preliminary | 1 | ≈ 0.76 pp. |
| C. Technical Details | 1 | ≈ 2.37 pp. |
| D. Implementation Details and Experimental | 4 | ≈ 0.59 pp. |
| E. More Evaluations | 4 | ≈ 0.74 pp. |
| F. More Ablation Studies | 5 | ≈ 3.02 pp. |
| G. Limitation and Discussion | 8 | ≈ 2.36 pp. |
- Table S1 · p. 1
- Figure S1 · p. 2
- Figure S2 · p. 3
- Figure S3 · p. 3
- Figure S4 · p. 3
- Figure S5 · p. 5
- Table S2 · p. 5
- Figure S6 · p. 6
- Figure S7 · p. 6
- Figure S8 · p. 6
- Figure S9 · p. 7
- Figure S10 · p. 7
- Table S3 · p. 7
- Figure S11 · p. 8
- Table S4 · p. 8
- Figure S12 · p. 8
- Figure S13 · p. 9
- Figure S14 · p. 9
- Figure S15 · p. 9
- Figure S16 · p. 9
- Figure S17 · p. 10
Argument and organization
- Problem and contribution
SmokeSVD separates matching the observed smoke projection from reconstructing plausible unseen structure and dynamics. This framing acknowledges the ambiguity of a single view rather than presenting front-view fidelity as sufficient validation.
- Argument sequence
The method first synthesizes a side view and then progressively reconstructs and refines novel views. Evaluation distinguishes real smoke, synthetic cases with physical fields, component ablations and failure/physical analyses.
- Evidence and verification
The results separate input-view and unseen-view measurements. Table 3 explicitly uses front and side inputs for a two-view comparison, whereas the central task is single-view; retaining this caption qualification prevents an unfair merged leaderboard.
Problem, claim and contribution
SmokeSVD separates matching the observed smoke projection from reconstructing plausible unseen structure and dynamics. This framing acknowledges the ambiguity of a single view rather than presenting front-view fidelity as sufficient validation.
Introduction; Sections 3–4 · Main paper · Additional source 2
Argument and section order
The method first synthesizes a side view and then progressively reconstructs and refines novel views. Evaluation distinguishes real smoke, synthetic cases with physical fields, component ablations and failure/physical analyses.
Sections 3.2–3.3 and 4 · Main paper · Additional source 2
Figures, captions and visual explanation
Figure 2 labels front, synthesized side and novel viewing directions, making which images are observations versus generated constraints inspectable. Figure 3 explains temporal diffusion conditioning, while the progressive refinement illustration supplies the view-order story.
Figures 2–4, PDF pp.4–5 · Main paper · Additional source 2
Tables, comparisons and evidence
The results separate input-view and unseen-view measurements. Table 3 explicitly uses front and side inputs for a two-view comparison, whereas the central task is single-view; retaining this caption qualification prevents an unfair merged leaderboard.
Tables 1–3; Figures 5–7, PDF p.6 · Main paper · Additional source 2
Appendix and reproducibility
The supplement expands mathematical notation, physical/temporal constraints, network details and additional real/synthetic studies. It also records alternatives and limitations; these details help interpret visual plausibility without pretending generated side views are measured truth.
Supplementary PDF pp.1–10 · Main paper · Additional source 2
Clarity and writing conventions
The main comparison prose admits when another method has better unseen-view quality and discusses the cost trade-off. It also states why projection error alone can fail to describe perceptual smoke quality.
Section 4 comparison discussion, PDF p.6 · Main paper · Additional source 2
What to reuse—and what not to infer
Transferable lesson: For ambiguous reconstruction, label observed and synthesized views, assess unseen structure separately from input fit, and state the number of input views next to each baseline comparison.
Boundary: A generated plausible side view is a prior, not an additional measurement. Two-view tables and single-view results must not be merged without their input conditions.
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 and references pp.9–11: abstract/framing, section sequence, figure/table captions, evaluation organization and conclusions; not an exhaustive technical verification.
- Appendix / supplement
- Writing-oriented supplement scan: all ten supplementary PDF pages, covering notation, temporal/physical modeling, network details, extra experiments and limitations. Video/code assets were not executed or assessed.
- Visually inspected PDF pages
- 4, 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.