Writing study: MAMMA: Markerless Accurate Multi-person Motion Acquisition

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Hanz Cuevas Velasquez, Anastasios Yiannakidis, Soyong Shin, Giorgio Becherini, Markus Höschle, Joachim Tesch, Taylor Obersat, Tsvetelina Alexiadis, Eni Halilaj, Michael J. Black · 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

12 physical pages

Before references / appendix

≈ 8.15 page-equivalents

Main figures / tables

7 / 5

Attached appendix figures / tables

0 / 0

Attached appendix span

Separate file recorded below; not in this PDF total

Before references / appendix: ≈ 67.92%References / intervening material: ≈ 32.08%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	≈ 1.08 pp.	9%
2. Related Work	2	≈ 0.83 pp.	6.92%
3. Dataset	3	≈ 0.96 pp.	8%
4. Method	4	≈ 1.53 pp.	12.75%
5. Experiments	6	≈ 2.63 pp.	21.92%
6. Conclusions	8	≈ 0.3 pp.	2.5%
References	9	≈ 3.85 pp.	32.08%

All detected figure and table locations

Main text

• Figure 1 · p. 1
• Figure 2 · p. 4
• Figure 3 · p. 4
• Figure 4 · p. 6
• Table 1 · p. 6
• Table 2 · p. 6
• Figure 5 · p. 7
• Figure 6 · p. 7
• Table 3 · p. 7
• Table 4 · p. 8
• Table 5 · 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 10 physical pages; 15 detected figures and 5 tables. These may include overlapping main-text material.

Located appendix only: starts on physical p.1; ≈ 9.88 page-equivalents, 15 figure labels and 5 table labels. This interval is distinct from the whole file.

Detected heading	PDF page	Estimated span
Supplementary Material	1	≈ 0.08 pp.
1. Landmark Networks Training Details	1	≈ 0.16 pp.
2. MammaSyn Dataset	1	≈ 0.57 pp.
3. Dataset Evaluation	1	≈ 0.45 pp.
4. MammaEval Dataset	2	≈ 0.49 pp.
5. Comparison with Harmony4D	2	≈ 0.52 pp.
6. Multiview Matching Evaluation	3	≈ 0.51 pp.
7. Optimization: Number of Cameras	3	≈ 0.8 pp.
8. Optimization: Stages Evaluation	4	≈ 0.16 pp.
9. Perceptual Study	4	≈ 1.01 pp.
10. Visualizing the Contact Probabilities	5	≈ 0.04 pp.
11. Capture Protocol	5	≈ 1.31 pp.
12. Usability and Cost	7	≈ 0.5 pp.
13. MAMMA Dance Release	7	≈ 0.09 pp.
14. More than Two People	7	≈ 0.54 pp.
15. Consumer-based Cameras	8	≈ 0.57 pp.
16. Limitations and Future Work	8	≈ 0.76 pp.
17. Supplementary Video Reveal	9	≈ 0.37 pp.
References	10	≈ 0.95 pp.

• Figure 1 · p. 1
• Table 1 · p. 1
• Figure 2 · p. 2
• Table 2 · p. 2
• Figure 3 · p. 3
• Table 3 · p. 4
• Figure 4 · p. 4
• Figure 6 · p. 4
• Figure 5 · p. 4
• Table 4 · p. 5
• Figure 7 · p. 5
• Figure 8 · p. 6
• Table 5 · p. 7
• Figure 9 · p. 7
• Figure 10 · p. 8
• Figure 11 · p. 8
• Figure 12 · p. 8
• Figure 13 · p. 8
• Figure 14 · p. 9
• Figure 15 · p. 9

Argument and organization

1. Problem and contribution

   Contrasts marker-based capture cost with the occlusion and interaction failures of markerless methods.

2. Argument sequence

   Explains synthetic training and real evaluation data before landmark prediction, fitting and correspondence, separating data novelty from model novelty.

3. Evidence and verification

   Evaluation progresses from landmarks to contact ROC, correspondence, mesh fitting and independent Vicon-marker comparison.

Problem, claim and contribution

Contrasts marker-based capture cost with the occlusion and interaction failures of markerless methods.

Introduction, pp. 1–2 · Main paper · Additional source 2

Argument and section order

Explains synthetic training and real evaluation data before landmark prediction, fitting and correspondence, separating data novelty from model novelty.

Sections 3–4, pp. 3–5 · Main paper · Additional source 2

Figures, captions and visual explanation

Figure 4 labels the network's distinct location, uncertainty, visibility and contact outputs with example overlays, instead of drawing one undifferentiated pose head.

Figure 4, p. 6 · Main paper · Additional source 2

Tables, comparisons and evidence

Evaluation progresses from landmarks to contact ROC, correspondence, mesh fitting and independent Vicon-marker comparison.

Sections 5.1–5.5 and Figure 6, pp. 6–8 · Main paper · Additional source 2

Appendix and reproducibility

Camera-count/runtime studies, capture protocol, usability/cost and consumer-camera examples make the acquisition requirements concrete.

Supplement Sections 7–16, pp. 3–8 · Main paper · Additional source 2

Clarity and writing conventions

Single-person and interaction data have explicit names and roles, preventing a broad accuracy claim from hiding the difficult overlap setting.

Sections 3 and 5, pp. 3–8 · Main paper · Additional source 2

What to reuse—and what not to infer

Transferable lesson: Validate each stage of a complex capture system and include an independent measurement reference, not only a final mesh gallery.

Boundary: Mask availability, camera synchronization and evaluation-marker independence matter; high-quality demos do not imply identical performance with arbitrary consumer capture.

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 Sections 1–6, PDF pp. 1–8; references pp. 9–12: abstract/framing, section sequence, figure/table captions, evaluation organization and conclusions; not an exhaustive technical verification.

Appendix / supplement

Writing-oriented supplement scan: Official ZIP-contained supplement Sections 1–17, pp. 1–10: data/training, capture rig, optimization/camera ablations, perception, Vicon protocol, cost and limits. 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.

• Official CVF open-access CVPR 2026 accepted-paper PDF · 12 pages
• Official CVF supplementary ZIP member PDF; see URL fragment for member · 10 pages