Writing study: HairCUP: Hair Compositional Universal Prior for 3D Gaussian Avatars

ICCV overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← ICCV 2025 oral writing corpus

Byungjun Kim, Shunsuke Saito, Giljoo Nam, Tomas Simon, Jason Saragih, Hanbyul Joo, Junxuan Li · 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

11 physical pages

Before references / appendix

≈ 8.1 page-equivalents

Main figures / tables

8 / 1

Attached appendix figures / tables

0 / 0

Attached appendix span

No boundary detected; not assumed zero

Before references / appendix: ≈ 73.64%References / intervening material: ≈ 26.36%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	≈ 0.74 pp.	6.73%
2. Related Work	2	≈ 0.63 pp.	5.73%
3. Method	3	≈ 3.51 pp.	31.91%
4. Experiments	6	≈ 2.05 pp.	18.64%
5. Conclusion	8	≈ 0.23 pp.	2.09%
Acknowledgments	9	≈ 0.06 pp.	0.55%
References	9	≈ 2.9 pp.	26.36%

All detected figure and table locations

Main text

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

Attached appendix

No recognized caption labels.

Argument and organization

1. 1 Introduction; 2 Related Work

   Identifies entangled face/hair priors and missing paired bald captures as coupled representation/data problems.

2. 3 Method

   Defines relightable primitives, separate identity/motion decoders, synthetic bald supervision and boundary-aware training.

3. 4 Experiments; 5 Conclusion

   Separates unseen-identity personalization, cross-identity transfer, interpolation and qualitative anchor/boundary ablations.

Problem, claim and contribution

Explains why paired synthetic bald supervision is needed to make compositionality learnable from ordinary haired capture subjects.

Abstract and Sections 1,3.3, physical pp.1–2,5 · Main paper

Argument and section order

The universal architecture and synthetic-data generation have separate subsections, followed by losses that distinguish composite, face-only and segmentation supervision.

Sections 3.2–3.4, pp.3–6 · Main paper

Figures, captions and visual explanation

The overview keeps face and hair pathways in separate colors and explicitly labels real composite, synthetic bald and segmentation targets; its caption distinguishes same-subject training from cross-subject testing.

Figure 2, p.4 · Main paper

Tables, comparisons and evidence

Personalization comparison averages eight unseen identities using front-view fully-lit head rotation; qualitative ablations separately vary anchor geometry and boundary segmentation treatment.

Section 4.1 and Table 1, p.7; Figures 7–8, p.8 · Main paper

Appendix and reproducibility

The preliminaries refer readers to the inherited relightable-avatar formulation; no detailed appendix roadmap was located in the main scan. Separate supplementary material remains outside the inspected scope.

Section 3.1, p.3; main argument through p.8 · Main paper

Clarity and writing conventions

The limitations name large hair motion, synthetic scalp color, fixed UV anchors and unobserved-scalp relighting as distinct unresolved cases.

Section 5, p.8 · Main paper

What to reuse—and what not to infer

Transferable lesson: Draw supervision provenance alongside compositional branches and use targeted visual ablations for boundary artifacts.

Boundary: Monocular personalization depends on a prior learned from specialized multi-camera/light captures. Synthetic bald targets are inferred rather than observed scalp truth; smooth examples do not establish complete disentanglement or arbitrary hair dynamics.

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 were checked. A separate supplementary source is linked in the inventory but was not retrieved or inspected; its contents are not included in the reading or visual-page counts.

Visually inspected PDF pages

4, 8

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 · 11 pages