Research map / 3D vision / Human pose estimation
A Stochastic Conditioning Scheme for Diverse Human Motion Prediction
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 Mix-and-Match stochastic conditioning instead of deterministic hidden-state/noise fusion | stochastic motion prediction · Human3.6M | Diversity: 0.26 -> 3.52; 1.70 -> 3.52; 0.48 -> 3.52 | Helped |
| Use Mix-and-Match stochastic conditioning instead of deterministic hidden-state/noise fusion | stochastic motion prediction · CMU Mocap | Diversity: 0.41 -> 2.63; 3.00 -> 2.63; 0.43 -> 2.63 | Mixed |
| Use Mix-and-Match stochastic conditioning instead of the compared stochastic baselines | stochastic motion prediction · Human3.6M | IS: 1.9±0.4 -> 7.3±1.4; 1.8±0.6 -> 7.3±1.4; 2.1±1.3 -> 7.3±1.4 | Helped |
| Use Mix-and-Match stochastic conditioning instead of the compared stochastic baselines | stochastic motion prediction · CMU Mocap | IS: 2.4±0.1 -> 9.0±1.7; 1.4±0.4 -> 9.0±1.7; 2.0±1.0 -> 9.0±1.7 | Helped |
| Use Mix-and-Match stochastic conditioning rather than the compared stochastic baselines | stochastic motion prediction · Human3.6M | Quality: 0.45, 0.13, 0.47 -> 0.42 | Mixed |
| Use Mix-and-Match stochastic conditioning rather than the compared stochastic baselines | stochastic motion prediction · CMU Mocap | Quality: 0.46, 0.18, 0.45 -> 0.46 | Mixed |
| Use Mix-and-Match stochastic conditioning rather than deterministic hidden-state/noise fusion | stochastic motion prediction · Human3.6M | ELBO: 0.51 -> 0.55; 2.08 -> 0.55; 0.61 -> 0.55 | Mixed |
| Use Mix-and-Match stochastic conditioning rather than deterministic hidden-state/noise fusion | stochastic motion prediction · CMU Mocap | ELBO: 0.25 -> 0.25; 1.93 -> 0.25; 0.24 -> 0.25 | Mixed |
| Concatenate pose with independent random vectors at each time step | Walker et al. stochastic motion predictor · Human3.6M | Quality: 0.13 | Hurt |
| Concatenate pose with independent random vectors at each time step | Walker et al. stochastic motion predictor · CMU Mocap | Quality: 0.18 | Hurt |
Closest papers by meaning
- Existence Is Chaos: Enhancing 3D Human Motion Prediction with Uncertainty Consideration (AAAI 2024 · similarity 0.90)
- Motion Diversification Networks (CVPR 2024 · similarity 0.88)
- Harmonizing Stochasticity and Determinism: Scene-responsive Diverse Human Motion Prediction (NeurIPS 2024 · similarity 0.88)
- FutureHuman3D: Forecasting Complex Long-Term 3D Human Behavior from Video Observations (CVPR 2024 · similarity 0.88)
- Stochastic Human Motion Prediction with Memory of Action Transition and Action Characteristic (CVPR 2025 · similarity 0.87)
- SimpliHuMoN: Simplifying Human Motion Prediction (NeurIPS 2026 · 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.