Research map / 3D vision / Human pose estimation
Diverse Human Motion Prediction Guided by Multi-level Spatial-Temporal Anchors
Sirui Xu, Yu-Xiong Wang, Liang-Yan Gui
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 anchor-based sampling to GSPS (ablation) | GSPS · Human3.6M (K=100) | APD: 14.751 -> 19.022 | Improved |
| add anchor-based sampling to GSPS (ablation) | GSPS · Human3.6M (K=100) | ADE: 0.369 -> 0.364 | Improved |
| add anchor-based sampling to IE-STGCN (ablation) | IE-STGCN · Human3.6M (K=100) | APD: 11.701 -> 14.554 | Improved |
| add anchor-based sampling to IE-STGCN (ablation) | IE-STGCN · Human3.6M (K=100) | ADE: 0.352 -> 0.344 | Improved |
| factorize anchors into 2 temporal x 25 spatial anchors instead of 50 original anchors (ablation) | STARS · Human3.6M (K=50) | APD: 16.974 -> 16.303 | Hurt |
| factorize anchors into 2 temporal x 25 spatial anchors instead of 50 original anchors (ablation) | STARS · Human3.6M (K=50) | ADE: 0.363 -> 0.356 | Improved |
| factorize anchors into 5 temporal x 10 spatial anchors instead of 50 original anchors (ablation) | STARS · HumanEva-I (K=50) | APD: 7.786 -> 6.199 | Hurt |
| add multi-level mechanism to 5 x 10 spatial-temporal anchors (ablation) | STARS · Human3.6M (K=50) | APD: 13.681 -> 15.884 | Improved |
| add multi-level mechanism to 5 x 10 spatial-temporal anchors (ablation) | STARS · Human3.6M (K=50) | ADE: 0.355 -> 0.358 | Hurt |
| add multi-level mechanism to 5 x 10 spatial-temporal anchors (ablation) | STARS · HumanEva-I (K=50) | ADE: 0.226 -> 0.217 | Improved |
| replace GSPS with STARS | GSPS · Human3.6M (K=50) | APD: 14.757 -> 15.884 | Improved |
| replace GSPS with STARS | GSPS · HumanEva-I (K=50) | ADE: 0.233 -> 0.217 | Improved |
Closest papers by meaning
- Generating Smooth Pose Sequences for Diverse Human Motion Prediction (ICCV 2021 · similarity 0.90)
- Multi-level Motion Attention for Human Motion Prediction (IJCV 2021 · similarity 0.90)
- Learning Trajectory Dependencies for Human Motion Prediction (ICCV 2019 · similarity 0.90)
- Progressively Generating Better Initial Guesses Towards Next Stages for High-Quality Human Motion Prediction (CVPR 2022 · similarity 0.90)
- GCNext: Towards the Unity of Graph Convolutions for Human Motion Prediction (AAAI 2024 · similarity 0.88)
- Motion Diversification Networks (CVPR 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.
