Writing study: MAGICIAN: Efficient Long-Term Planning with Imagined Gaussians for Active Mapping

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Shiyao Li, Antoine Guédon, Shizhe Chen, Vincent Lepetit · 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: Efficiency, inference & systems (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

≈ 8.12 page-equivalents

Main figures / tables

8 / 5

Attached appendix figures / tables

0 / 0

Attached appendix span

Separate file recorded below; not in this PDF total

Before references / appendix: ≈ 81.2%References / intervening material: ≈ 18.8%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	2	≈ 0.76 pp.	7.6%
2. Related work	2	≈ 0.93 pp.	9.3%
3. Method	3	≈ 2.17 pp.	21.7%
4. Experiments	5	≈ 2.93 pp.	29.3%
5. Conclusion	8	≈ 0.21 pp.	2.1%
Acknowledgements	9	≈ 0.08 pp.	0.8%
References	9	≈ 1.88 pp.	18.8%

All detected figure and table locations

Main text

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

Located appendix only: starts on physical p.1; ≈ 7.93 page-equivalents, 3 figure labels and 4 table labels. This interval is distinct from the whole file.

Detected heading	PDF page	Estimated span
Supplementary Material	1	≈ 0.08 pp.
A. Method	1	≈ 1.9 pp.
B. Experiments	3	≈ 3.19 pp.
C. Failure Case and Analysis	6	≈ 0.81 pp.
References	7	≈ 1.95 pp.

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

Argument and organization

1. Method §§3.1–3.5

   Define the exploration objective, derive coverage gain, introduce occupancy/Gaussians and then the planner.

2. Experiments §§4.1–4.3

   State embodiment configurations, compare mapping quality and analyze search horizon/replanning choices.

Problem, claim and contribution

The opening distinguishes active mapping with known poses from passive SLAM, then ties greedy viewpoints to inefficient reconstruction coverage.

§1, PDF p. 2 · Main paper · Additional source 2

Argument and section order

The coverage-gain objective precedes the representation and planner, making fast rendering a means of enabling long-horizon search rather than an unrelated graphics component.

§§3.2–3.5, PDF pp. 3–5 · Main paper · Additional source 2

Figures, captions and visual explanation

Figure 2 closes the loop from RGB-D observations to occupancy, novelty fields, candidate trajectories and execution. The caption distinguishes planning depth from the shorter action prefix actually executed.

Figure 2, PDF p. 3 · Main paper · Additional source 2

Tables, comparisons and evidence

Table 1 exposes different robot action spaces, camera height, resolution and step budgets; reconstruction quality is evaluated from a fixed 100-image trajectory sample with a common reconstruction method.

Table 1 and Figure 5, PDF p. 6; Figure 6, PDF p. 7 · Main paper · Additional source 2

Appendix and reproducibility

The supplement adds per-scene coverage/AUC, variance across starting poses, sampling-density sensitivity and failure analysis of poor early occupancy prediction.

Supplement §§A–C, PDF pp. 1–8; Tables 6–9 and Figure 9 · Main paper · Additional source 2

Clarity and writing conventions

Beam width, look-ahead depth and replanning frequency have separate analyses, so long-term planning quality is not confused with an unspecified larger compute budget.

Figures 7–8, PDF p. 8 · Main paper · Additional source 2

What to reuse—and what not to infer

Transferable lesson: For a planning system, show the closed loop and execution horizon explicitly, then report both exploration efficiency and reconstruction quality under disclosed embodiment budgets.

Boundary: Strong results with known poses and a learned occupancy prior are not demonstrated robustness to arbitrary real robotic localization errors; early incomplete geometry remains a reported failure mode.

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

Manuscript-wide writing scan of PDF pp. 1–10: framing, argument sequence, section organization, figure/table captions, evaluation interpretation and limitations; reference sections identified. This is a writing-focused scan, not an exhaustive technical proof or reproduction audit. Main PDF pages 3 and 6 were rendered and visually inspected.

Appendix / supplement

Writing-focused scan of the separate supplementary PDF, pp. 1–8, including its organization, captions, protocol/evidence detail and limitation material where present; proofs and code listings were inspected for exposition, not independently verified or executed. Supplement §§A–C, PDF pp. 1–8; Tables 6–9 and Figure 9. Linked video/audio demonstrations and non-PDF archive contents were not 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 CVPR 2026 CVF open-access accepted manuscript; retrieved 2026-09-26 · 10 pages
• Official CVPR 2026 CVF supplementary PDF; retrieved 2026-09-26 · 8 pages