Research map / Video / Video segmentation and tracking
STaR: Seamless Spatial-Temporal Aware Motion Retargeting with Penetration and Consistency Constraints
Xiaohang Yang, Qing Wang, Jiahao Yang, Gregory Slabaugh, Shanxin Yuan
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 spatio-temporal model (ablation) | frame-by-frame baseline · Mixamo | Curv: 66.35 -> 7.87 | Improved |
| add spatio-temporal model (ablation) | frame-by-frame baseline · Mixamo | Pen%: 8.74 -> 9.45 | Hurt |
| add limb penetration constraint L_lp (ablation) | ST model · Mixamo | Pen%: 9.45 -> 8.20 | Improved |
| add basic smoothness loss L_t (ablation) | ST + L_lp · Mixamo | Pen%: 8.20 -> 8.40 | No clear effect |
| add dense shape representation (ablation) | ST + L_lp + L_t · Mixamo | Curv: 35.74 -> 21.29 | Improved |
| add temporal consistency constraint L_tc (ablation) | ST + L_lp + DSR + L_t · Mixamo | Curv: 21.29 -> 10.66 | Improved |
| use larger weight on L_lp (ablation) | ST + L_lp + DSR + L_t · Mixamo | Pen%, MSE: Pen% 3.63, MSE 0.2206 | Mixed |
| reduce sampled vertices for L_lp from 4,000 to 500 (ablation) | STaR · Mixamo | Pen%: 7.99 -> 8.49 | No clear effect |
| add skeleton info to DSR (ablation) | STaR · Mixamo | Pen%: 7.99 -> 8.22 | Hurt |
| replace DSR with skeleton info only (ablation) | STaR · Mixamo | Pen%: 7.99 -> 9.42 | Hurt |
| separate point cloud per joint instead of shared DSR (ablation) | STaR · Mixamo | Pen%: 7.99 -> 13.45 | Hurt |
| add global motion decoder (ablation) | baseline height-normalized global motion · Mixamo | MSE: 1.7770 -> 1.8708 | Hurt |
Closest papers by meaning
- Skinned Motion Retargeting With Residual Perception of Motion Semantics & Geometry (CVPR 2023 · similarity 0.87)
- Semantics-aware Motion Retargeting with Vision-Language Models (CVPR 2024 · similarity 0.87)
- MoCaNet: Motion Retargeting In-the-Wild via Canonicalization Networks (AAAI 2022 · similarity 0.85)
- Skinned Motion Retargeting with Dense Geometric Interaction Perception (NeurIPS 2024 · similarity 0.85)
- Self-Consistency for LLM-Based Motion Trajectory Generation and Verification (CVPR 2026 · similarity 0.82)
- Dynamic Motion Blending for Versatile Motion Editing (CVPR 2025 · similarity 0.82)
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.
