Research map / Video / Video generation
SMGDiff: Soccer Motion Generation using Diffusion Probabilistic Models
Hongdi Yang, Chengyang Li, Zhenxuan Wu, Gaozheng Li, Jingya Wang, Jingyi Yu, Zhuo Su, Lan Xu
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 |
|---|---|---|---|
| raise acceleration threshold from 0.5 to 2 m/s^2 (ablation) | SMGDiff contact guidance · Soccer-X ball-foot contact detection | accuracy: 56.66% -> 91.47% | Improved |
| raise acceleration threshold from 2 to 3 m/s^2 (ablation) | SMGDiff contact guidance · Soccer-X ball-foot contact detection | accuracy: 91.47% -> 83.31% | Hurt |
| raise acceleration threshold from 1 to 1.5 m/s^2 (ablation) | SMGDiff contact guidance · Soccer-X ball-foot contact detection | accuracy: 67.13% -> 79.12% | Improved |
| set penalty term w_d to 2 instead of 1 (ablation) | SMGDiff contact guidance · Soccer-X contact joint identification | accuracy: 88.31% -> 94.43% | Improved |
| set penalty term w_d to 2 instead of 3 (ablation) | SMGDiff contact guidance · Soccer-X contact joint identification | accuracy: 87.01% -> 94.43% | Improved |
| set penalty term w_d to 2 instead of 2.5 (ablation) | SMGDiff contact guidance · Soccer-X contact joint identification | accuracy: 91.41% -> 94.43% | Improved |
Closest papers by meaning
- Sports-Traj: A Unified Trajectory Generation Model for Multi-Agent Movement in Sports (ICLR 2025 · similarity 0.83)
- Event2Tracking: Reconstructing Multi-Agent Soccer Trajectories Using Long-Term Multimodal Context (AAAI 2025 · similarity 0.82)
- Unified Uncertainty-Aware Diffusion for Multi-Agent Trajectory Modeling (CVPR 2025 · similarity 0.82)
- Rethinking Diffusion for Text-Driven Human Motion Generation: Redundant Representations, Evaluation, and Masked Autoregression (CVPR 2025 · similarity 0.80)
- MotionMix: Weakly-Supervised Diffusion for Controllable Motion Generation (AAAI 2024 · similarity 0.80)
- Towards Universal Soccer Video Understanding (CVPR 2025 · similarity 0.80)
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.
