Writing study: Monocular Open Vocabulary Occupancy Prediction for Indoor Scenes
← CVPR 2026 oral writing corpus
Changqing Zhou, Yueru Luo, Han Zhang, Zeyu Jiang, Changhao Chen · Official CVPR 2026 oral record
Figure, table and section ledger
Editorial contribution type: Training, objectives & optimization (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
- ≈ 8.09 page-equivalents
- Main figures / tables
- 5 / 4
- 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.
All detected figure and table locations
Main text
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 2 physical pages; 4 detected figures and 2 tables. These may include overlapping main-text material.
Located appendix only: starts on physical p.1; ≈ 1.94 page-equivalents, 4 figure labels and 2 table labels. This interval is distinct from the whole file.
| Detected heading | PDF page | Estimated span |
|---|---|---|
| Supplementary Material | 1 | ≈ 0.02 pp. |
| 6. Experimental Results | 1 | ≈ 0.83 pp. |
| 7. Limitations | 1 | ≈ 1.09 pp. |
Read-only audit 2026-09-26: restored actual supplementary section headings from cached PDF layout, excluding table-of-contents entries and retaining physical PDF page coordinates. This measures reading-order spans, not occupied text area.
Argument and organization
- Geometry-only indoor problem
Separate geometric occupancy from semantic labeling before introducing the common Gaussian representation.
- Two mechanisms and matched tests
Explain Poisson aggregation and temperature decay, then test each under explicitly separated supervision settings.
Problem, claim and contribution
The introduction defines the missing supervision precisely: binary occupancy remains available, whereas semantic voxel annotations are excluded. This prevents open-vocabulary from being confused with fully unsupervised reconstruction.
Sections 1 and 3.1 · Main paper · Additional source 2
Argument and section order
The method progresses from the representation to occupancy aggregation, semantic alignment and the joint loss, giving each experiment a corresponding mechanism.
Sections 3.1–3.5 · Main paper · Additional source 2
Figures, captions and visual explanation
The overview separates geometry learning, feature rendering and inference-only text similarity; the caption names the shared representation rather than leaving arrows unexplained.
Figure 2, PDF p. 4; caption scan · Main paper · Additional source 2
Tables, comparisons and evidence
Tables 2–4 distinguish closed/open-vocabulary settings using labeled groups and shading; the temperature table retains collapsed runs, and the speed table names hardware and warm-up protocol.
Tables 2–4, PDF p. 7 · Main paper · Additional source 2
Appendix and reproducibility
The supplement explains furniture/objects and sofa/chair ambiguity, then adds view-count and resolution/FPS tests instead of only further attractive renderings.
Supplement pp. 1–2, Tables 5–6, Section 7 · Main paper · Additional source 2
Clarity and writing conventions
IoU and mIoU remain separate throughout, allowing the discussion to acknowledge stronger geometry without implying semantic performance exceeds fully annotated models.
Section 4.3 and Table 1 · Main paper · Additional source 2
What to reuse—and what not to infer
Transferable lesson: Pair each proposed training intervention with its own failure-mode ablation, and display supervision regimes explicitly in comparison tables.
Boundary: The evaluated taxonomy is closed even when training is open vocabulary; arbitrary-text qualitative examples do not establish exhaustive category generalization. Geometry labels remain required.
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 main PDF pp. 1–8: framing, section structure, method exposition, figure/table captions, evaluation organization and conclusions; not an exhaustive technical proof audit or reproduction.
- Appendix / supplement
- Separate supplement pp. 1–2 scanned: grounding examples, class ambiguity, rendered-view and image-size ablations, and explicit limitations.
- Visually inspected PDF pages
- 3, 7
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.