Research map / 3D vision / Human pose estimation
MimiCAT: Mimic with Correspondence-Aware Cascade-Transformer for Category-Free 3D Pose Transfer
Zenghao Chai, Chen Tang, Yongkang Wong, Xulei Yang, Mohan Kankanhalli
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add rotation blending (Eq. 4) instead of equal weighted sum of quaternions (ablation) | MimiCAT · Mixamo H2H | PMD (x100): 4.439 -> 3.570 | Improved |
| add pose prior regularization (Eq. 8) (ablation) | MimiCAT · Cross-category transfer | PMD (x100): 4.655 -> 4.264 | Improved |
| add pose prior regularization (Eq. 8) (ablation) | MimiCAT · Mixamo H2H | PMD (x100): 4.161 -> 3.570 | Improved |
| use text-based correspondence supervision instead of hierarchical correspondence (ablation) | MimiCAT · Cross-category transfer | PMD (x100): 4.612 -> 4.264 | Improved |
| use text-based correspondence supervision instead of hierarchical correspondence (ablation) | MimiCAT · Mixamo H2H | ELS: 0.919 -> 0.923 | Improved |
| add MimiCAT components (soft correspondence, shape-aware transformer) vs SFPT | SFPT · Cross-category transfer | PMD (x100): 4.312 -> 4.264 | Improved |
| add MimiCAT vs SFPT on humanoid-to-humanoid | SFPT · Mixamo H2H | PMD (x100): 3.616 -> 3.570 | Improved |
| use one-to-one correspondence baselines (NPT) across categories | NPT · Cross-category transfer | ELS: 0.842 (H2H) -> 0.260 (CCT) | Hurt |
Closest papers by meaning
- Skeleton-Free Pose Transfer for Stylized 3D Characters (ECCV 2022 · similarity 0.89)
- PoseMaster: A Unified 3D Native Framework for Stylized Pose Generation (CVPR 2026 · similarity 0.87)
- Free-viewpoint Human Animation with Pose-correlated Reference Selection (CVPR 2025 · similarity 0.86)
- TrioPose: Native Triple-Stream Diffusion Transformers for Pose-Guided Text-to-Image Generation (NeurIPS 2026 · similarity 0.85)
- Disentangled Pose and Appearance Guidance for Multi-Pose Generation (CVPR 2025 · similarity 0.85)
- Unifying Correspondence Pose and NeRF for Generalized Pose-Free Novel View Synthesis (CVPR 2024 · similarity 0.85)
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.
