Research map / Video / Video generation
High-Fidelity Neural Human Motion Transfer From Monocular Video
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use explicit shape and internal clothing-structure conditioning rather than pose-only image synthesis | Human motion transfer · Self-reenactment test set | SSIM: 0.9288 -> 0.9358 | Helped |
| Use explicit shape and internal clothing-structure conditioning rather than pose-only image synthesis | Human motion transfer · Self-reenactment test set | LPIPS: 0.0401 -> 0.0289 | Helped |
| Use explicit shape and internal clothing-structure conditioning rather than pose-only image synthesis | Human motion transfer · Self-reenactment test set | FID: 13.5421 -> 17.9573 | Hurt |
| Add shape conditioning (ablation) | P pose-only appearance network · Swinging dress self-reenactment | SSIM: 0.8959 -> 0.9045 | Helped |
| Add shape conditioning (ablation) | P pose-only appearance network · Swinging dress self-reenactment | LPIPS: 0.044 -> 0.036 | Helped |
| Add shape conditioning (ablation) | P pose-only appearance network · Swinging dress self-reenactment | FID: 15.046 -> 12.282 | Helped |
| Add internal clothing-structure conditioning (ablation) | PS shape-conditioned network · Swinging dress self-reenactment | LPIPS: 0.036 -> 0.035 | Helped |
| Add internal clothing-structure conditioning (ablation) | PS shape-conditioned network · Swinging dress self-reenactment | FID: 12.282 -> 12.176 | Helped |
| Add internal clothing-structure conditioning (ablation) | PS shape-conditioned network · Swinging dress self-reenactment | SSIM: 0.9045 -> 0.9046 | Helped |
| Add recurrent feedback of previous network outputs (ablation) | PSS motion-transfer framework · Swinging dress self-reenactment | temporal consistency | Helped |
| Add recurrent feedback of previous network outputs (ablation) | PSS motion-transfer framework · Swinging dress self-reenactment | SSIM: 0.9045 -> 0.9046 | No clear effect |
| Use the proposed method instead of EDN | Video motion transfer · Liu et al. tight-clothing dataset, expert preference study | votes: 103 vs 76 | Hurt |
Closest papers by meaning
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.89)
- DLCA-Recon: Dynamic Loose Clothing Avatar Reconstruction from Monocular Videos (AAAI 2024 · similarity 0.88)
- HumanDreamer: Generating Controllable Human-Motion Videos via Decoupled Generation (CVPR 2025 · similarity 0.88)
- Pursuing Temporal-Consistent Video Virtual Try-On via Dynamic Pose Interaction (CVPR 2025 · similarity 0.87)
- MimicMotion: High-Quality Human Motion Video Generation with Confidence-aware Pose Guidance (ICML 2025 · similarity 0.87)
- DisPose: Disentangling Pose Guidance for Controllable Human Image Animation (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.