Writing study: SoccerMaster: A Vision Foundation Model for Soccer Understanding

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Haolin Yang, Jiayuan Rao, Haoning Wu, Weidi Xie · 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

≈ 8.09 page-equivalents

Main figures / tables

3 / 7

Attached appendix figures / tables

0 / 0

Attached appendix span

Separate file recorded below; not in this PDF total

Before references / appendix: ≈ 67.42%References / intervening material: ≈ 32.58%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	≈ 0.82 pp.	6.83%
2. Related Work	2	≈ 0.3 pp.	2.5%
3. Task Specifications	2	≈ 0.75 pp.	6.25%
4. Pretraining Dataset	3	≈ 1.25 pp.	10.42%
5. Models	4	≈ 2.39 pp.	19.92%
6. Experiments	7	≈ 1.52 pp.	12.67%
7. Conclusion	8	≈ 0.25 pp.	2.08%
Acknowledgments	9	≈ 0.04 pp.	0.33%
References	9	≈ 3.91 pp.	32.58%

All detected figure and table locations

Main text

• Figure 1 · p. 1
• Figure 2 · p. 4
• Table 1 · p. 4
• Table 2 · p. 4
• Figure 3 · p. 5
• Table 3 · p. 7
• Table 4 · p. 7
• Table 5 · p. 8
• Table 6 · p. 8
• Table 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 11 physical pages; 3 detected figures and 5 tables. These may include overlapping main-text material.

Located appendix only: starts on physical p.2; ≈ 9.91 page-equivalents, 3 figure labels and 5 table labels. This interval is distinct from the whole file.

Detected heading	PDF page	Estimated span
A. Pretraining Dataset Construction and Integration	2	≈ 2.63 pp.
B. More Implementation Details	4	≈ 1.49 pp.
C. Details of Camera Calibration Metrics	6	≈ 0.13 pp.
D. Ablation on Multi-task Pretraining	6	≈ 0.28 pp.
E. Qualitative Results	6	≈ 0.78 pp.
F. Limitations & Future Works	7	≈ 3.7 pp.
References	11	≈ 0.91 pp.

• Table 1 · p. 4
• Table 2 · p. 5
• Table 3 · p. 5
• Table 4 · p. 5
• Table 5 · p. 6
• Figure 1 · p. 8
• Figure 2 · p. 9
• Figure 3 · p. 10

Argument and organization

1. Task specifications and data §§3–4

   Define each output and evaluate the annotation pipeline before presenting the model.

2. Model and experiments §§5–6

   Separate visual encoding, task losses and downstream adaptations, then benchmark and ablate data effects.

Problem, claim and contribution

The core tension is fine-grained spatial perception versus high-level semantic understanding in soccer, and the contributions pair a shared model with the annotations needed to train it.

Abstract and introduction, PDF pp. 1–2 · Main paper · Additional source 2

Argument and section order

Task specifications define detection, pitch geometry, classification, alignment and downstream tasks before the training architecture, making a broad model claim a list of testable outputs.

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

Figures, captions and visual explanation

Data curation and model architecture receive separate full diagrams; the latter distinguishes spatial/temporal encoding from pretraining and downstream task heads.

Figure 2, PDF p. 4; Figure 3, PDF p. 5 · Main paper · Additional source 2

Tables, comparisons and evidence

The pretraining comparison freezes encoders with trainable task heads and marks automatic-data inclusion. Camera calibration explicitly distinguishes zero-shot from fine-tuned results, while tracking and commentary have separate tables.

Tables 3–4, PDF p. 7; Tables 6–7, PDF p. 8 · Main paper · Additional source 2

Appendix and reproducibility

A contents-indexed supplement provides source/split integration, component learning rates, loss weights, augmentations and multitask ablations; limitations identify jersey-number imbalance and prediction mistakes.

Supplement §§A–F, PDF pp. 1–11; Tables 1–5 · Main paper · Additional source 2

Clarity and writing conventions

Bold task labels organize the several losses and outputs, preventing equations for detection, pitch registration and alignment from appearing as an undifferentiated objective list.

§§5.3–5.4, PDF p. 6 · Main paper · Additional source 2

What to reuse—and what not to infer

Transferable lesson: For a domain foundation model, define the task contract first, audit automatic annotations independently and distinguish pretraining probes from downstream adaptation.

Boundary: Unified representation does not mean one identical head or zero-shot protocol for every task; preserve the task-specific heads, training data and calibration footnotes.

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. 1–12: 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. 1–11, 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–F, PDF pp. 1–11; Tables 1–5. 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 · 12 pages
• Official CVPR 2026 CVF supplementary PDF; retrieved 2026-09-26 · 11 pages