Writing study: TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models

ICCV overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← ICCV 2025 oral writing corpus

Mark Yu, Wenbo Hu, Jinbo Xing, Ying Shan · 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

12 physical pages

Before references / appendix

≈ 8.05 page-equivalents

Main figures / tables

9 / 3

Attached appendix figures / tables

0 / 0

Attached appendix span

No boundary detected; not assumed zero

Before references / appendix: ≈ 67.08%References / intervening material: ≈ 32.92%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	≈ 0.8 pp.	6.67%
2. Related Work	2	≈ 0.68 pp.	5.67%
3. Method	3	≈ 2.47 pp.	20.58%
4. Experiments	5	≈ 3.01 pp.	25.08%
5. Conclusion	8	≈ 0.23 pp.	1.92%
References	9	≈ 3.95 pp.	32.92%

All detected figure and table locations

Main text

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

Attached appendix

No recognized caption labels.

Argument and organization

1. 1 Introduction; 2 Related Work

   Frames camera redirection in casual dynamic video as a joint geometry and completion problem.

2. 3 Method

   Separates point-cloud reprojection, dual-stream conditional diffusion and hybrid training-data construction.

3. 4 Experiments; 5 Conclusion

   Uses reference-based multiview evaluation and a separate no-reference video benchmark, followed by architecture/data ablations and depth-induced failure.

Problem, claim and contribution

Distinguishes deterministic view transformation from stochastic filling of unseen content, giving each a different role in the method.

Abstract and Sections 1,3.2, physical pp.1–3 · Main paper

Argument and section order

The method explains geometric conditioning before the reference branch and then explains why double reprojection makes monocular video usable for training.

Sections 3.2–3.4, pp.3–5 · Main paper

Figures, captions and visual explanation

The overview gives separate colors to view/reference/text tokens and inherited/added blocks, exposing exactly where the added reference stream enters the pretrained diffusion pipeline.

Figure 2, p.3 · Main paper

Tables, comparisons and evidence

The quantitative page separates five-scene reference metrics from VBench scores over 100 real and 60 generated videos; accompanying prose explains why plausible disocclusions can still score poorly in PSNR.

Tables 1–2 and Sections 4.2–4.3, p.6 · Main paper

Appendix and reproducibility

The first figure points to the project page for videos; this scan checks the static manuscript only and does not establish temporal fidelity from those unviewed videos or any separate supplement.

Figure 1, p.1; main reference boundary after p.8 · Main paper

Clarity and writing conventions

The limitations name 360-degree motion and unreliable depth, with a concrete dog/glass collision rather than an abstract statement about future improvements.

Section 4.5 and Figure 9, p.8 · Main paper

What to reuse—and what not to infer

Transferable lesson: Give incompatible evaluation regimes separate tables and explicitly explain what a metric cannot reward in a generative task.

Boundary: VBench quality and selected static frames do not certify geometric correctness or unrestricted trajectory control. Empirically chosen intrinsics and depth errors remain part of the input contract.

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 and the reference boundary were checked. No separate supplementary link appears in this dated CVF inventory; this does not establish that no supplementary material exists elsewhere.

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