Research map / 3D vision / 3D generation
Context-Aware 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 |
|---|---|---|---|
| Add learned context interactions to human motion prediction | RNN on WBHM, all actions · WBHM, all actions | mean Euclidean error (mm) at 2 s: 191 -> 177 | Helped |
| Add object motion prediction and learned interactions | RNN object motion prediction on WBHM · WBHM, all actions | mean Euclidean error (mm) at 2 s: 135 -> 121 | Helped |
| Add learned interactions between people | RNN on CMU MoCap · CMU MoCap | mean Euclidean error (mm) at 2.0 s: 477 -> 435 | Helped |
| Add learned context interactions | RNN human motion prediction with noisy observations · WBHM, noisy input (25 mm) | mean Euclidean error (mm) at 2 s: 185 -> 166 | Helped |
| Add learned context interactions | RNN human motion prediction with noisy observations · WBHM, noisy input (50 mm) | mean Euclidean error (mm) at 2 s: 192 -> 175 | Helped |
| Replace learned interactions with distance-threshold interactions (ablation) | C-RNN+LI human motion prediction · WBHM, all actions | mean Euclidean error (mm) at 2 s: 177 -> 196 | Hurt |
| Replace learned interactions with distance-threshold interactions in the object-motion model (ablation) | C-RNN+OMP+LI human motion prediction · WBHM, all actions | mean Euclidean error (mm) at 2 s: 178 -> 197 | Hurt |
| Add object motion prediction to the learned-interaction human model (ablation) | C-RNN+LI · WBHM, all actions | mean Euclidean error (mm) at 2 s: 177 -> 178 | Hurt |
| Add learned interactions to the object motion prediction model (ablation) | C-RNN+OMP · WBHM, all actions | mean Euclidean error (mm) at 2 s: 125 -> 121 | Helped |
| Add learned interactions to human motion prediction | RNN with noise-free observations · WBHM, noise-free input | mean Euclidean error (mm) at 2 s: 179 -> 161 | Helped |
Closest papers by meaning
- Action-guided 3D Human Motion Prediction (NeurIPS 2021 · similarity 0.90)
- On Human Motion Prediction Using Recurrent Neural Networks (CVPR 2017 · similarity 0.90)
- A Neural Temporal Model for Human Motion Prediction (CVPR 2019 · similarity 0.89)
- Expressive Forecasting of 3D Whole-Body Human Motions (AAAI 2024 · similarity 0.87)
- Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs (NeurIPS 2026 · similarity 0.87)
- GCNext: Towards the Unity of Graph Convolutions for Human Motion Prediction (AAAI 2024 · 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.