Research map / Probabilistic and scientific modeling / Variational and generative sampling
Bayesian Adversarial Human Motion Synthesis
Rui Zhao, Hui Su, Qiang Ji
Check 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 |
|---|---|---|---|
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM motion synthesis · CMU Motion Capture | Inception Score: 2.77±0.08 -> 2.86±0.10 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM motion synthesis · Berkeley MHAD | Inception Score: 6.19±0.38 -> 6.49±0.23 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM motion synthesis · CMU Motion Capture | MMD: 3.98±0.23 -> 2.41±0.35 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM motion synthesis · Berkeley MHAD | MMD: 67.26±0.21 -> 48.70±0.11 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM data restoration · CMU Motion Capture, Joint set A | PCC: 0.86±0.12 -> 0.92±0.06 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM data restoration · CMU Motion Capture, Joint set B | PCC: 0.78±0.22 -> 0.85±0.15 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM data restoration · Berkeley MHAD, Joint set A | PCC: 0.53±0.42 -> 0.68±0.25 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM data restoration · Berkeley MHAD, Joint set B | PCC: 0.67±0.36 -> 0.73±0.16 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM data restoration · CMU Motion Capture, Joint set A | MSE: 0.28±0.15 -> 0.11±0.05 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM data restoration · CMU Motion Capture, Joint set B | MSE: 0.24±0.79 -> 0.14±0.03 | Hurt |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM data restoration · Berkeley MHAD, Joint set A | MSE: 0.13±0.38 -> 0.12±0.29 | Improved |
| Use Bayesian posterior parameter samples instead of a single MAP estimate (ablation) | BH-HSMM data restoration · Berkeley MHAD, Joint set B | MSE: 0.21±0.31 -> 0.23±0.37 | Hurt |
Closest papers by meaning
- An Adversarial Hierarchical Hidden Markov Model for Human Pose Modeling and Generation (AAAI 2018 · similarity 0.90)
- Scene-Aware Generative Network for Human Motion Synthesis (CVPR 2021 · similarity 0.88)
- Pose Guided Human Video Generation (ECCV 2018 · similarity 0.87)
- Generating Smooth Pose Sequences for Diverse Human Motion Prediction (ICCV 2021 · similarity 0.86)
- Autoregressive Motion Generation with Gaussian Mixture-Guided Latent Sampling (NeurIPS 2025 · similarity 0.86)
- Deterministic-to-Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis (CVPR 2025 · 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.
