Writing study: MikuDance: Animating Character Art with Mixed Motion Dynamics

ICCV overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← ICCV 2025 oral writing corpus

Jiaxu Zhang, Xianfang Zeng, Xin Chen, Wei Zuo, Gang Yu, Zhigang Tu · 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

≈ 8.12 page-equivalents

Main figures / tables

11 / 1

Attached appendix figures / tables

0 / 0

Attached appendix span

No boundary detected; not assumed zero

Before references / appendix: ≈ 73.82%References / intervening material: ≈ 26.18%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	≈ 1.23 pp.	11.18%
2. Related Work	2	≈ 0.63 pp.	5.73%
3. Method	3	≈ 2.37 pp.	21.55%
4. Experiments	5	≈ 3.26 pp.	29.64%
5. Conclusions	8	≈ 0.31 pp.	2.82%
References	9	≈ 2.88 pp.	26.18%

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
• Figure 6 · p. 6
• Figure 7 · p. 7
• Figure 8 · p. 7
• Figure 9 · p. 7
• Figure 10 · p. 8
• Table 1 · p. 8
• Figure 11 · p. 8

Attached appendix

No recognized caption labels.

Argument and organization

1. 1 Introduction; 2 Related Work

   Separates dynamic scene motion from reference/pose shape-scale mismatch.

2. 3 Method

   Unifies scene and character guidance in pixel space, injects motion-adaptive normalization and defines two-stage mixed-source training.

3. 4 Experiments; 5 Conclusions

   Organizes high-motion, misalignment and style demonstrations before quantitative ablations, user study and background-distortion limits.

Problem, claim and contribution

The two challenges map to mixed motion modeling and mixed-control diffusion rather than a general claim that ordinary human animation transfers unchanged to art.

Sections 1–2, physical pp.1–2 · Main paper

Argument and section order

A separate training subsection explains stylized frame pairs first and character/non-character video mixtures second, including which modules are frozen.

Section 3.4 and Figure 4, p.5 · Main paper

Figures, captions and visual explanation

The overview marks body guidance as required and face/hand/scene guidance as optional, and expands the motion-adaptive normalization branch.

Figure 2, p.3 · Main paper

Tables, comparisons and evidence

The table includes both an original and MMD-fine-tuned Animate Anyone baseline, while the ablation grid separates architecture removal from alternate motion representations.

Table 1 and Figure 10, p.8 · Main paper

Appendix and reproducibility

More art-style results are routed to the appendix; this scan viewed only static main-paper panels, not animation videos or supplemental demonstrations.

Figure 9, p.7 · Main paper

Clarity and writing conventions

The quantitative evaluation distinguishes 100 held-out MMD videos with first-frame references from qualitative art synthesized with SDXL-Neta.

Experiment datasets, p.5 · Main paper

What to reuse—and what not to infer

Transferable lesson: For animation systems, distinguish reference-domain construction, paired quantitative tests and open-ended qualitative examples, and include a domain-fine-tuned baseline.

Boundary: Generated-art examples are not a representative audit of all artists’ styles. Camera/depth/pose extraction contribute upstream assumptions; first-frame-reference metrics differ from arbitrary-reference animation, and static page checks cannot verify long-video temporal quality or the reported preference percentage.

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

3, 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