Research map / 3D vision / Human pose estimation
Pose Forecasting in Industrial Human-Robot Collaboration
Alessio Sampieri, Guido Maria D’Amely di Melendugno, Andrea Avogaro, Federico Cunico, Francesco Setti, Geri Skenderi, Marco Cristani, Fabio Galasso
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 |
|---|---|---|---|
| use space-time separable adjacency (STS-GCN) instead of plain GCN (ablation) | GCN · Human3.6M (1000 ms) | MPJPE (mm): 123.2 -> 117.0 | Improved |
| add a 5th layer to STS-GCN (ablation) | STS-GCN · Human3.6M (1000 ms) | MPJPE (mm): 117.0 -> 115.9 | Improved |
| add a 6th layer to STS-GCN (ablation) | STS-GCN · Human3.6M (1000 ms) | MPJPE (mm): 115.9 -> 116.1 | Hurt |
| add MLP layers between GCN layers (5+5) (ablation) | STS-GCN · Human3.6M (1000 ms) | MPJPE (mm): 115.9 -> 125.2 | Hurt |
| use depth-wise separable convolutions with MLP layers (5+5) (ablation) | STS-GCN · Human3.6M (1000 ms) | MPJPE (mm): 117.0 -> 114.8 | Improved |
| add depth-wise separable convolution to STS-GCN, parameters (ablation) | STS-GCN · Human3.6M (1000 ms) | parameters (K): 57.6 -> 70.0 | Hurt |
| sparsify adjacency with Sparse-GCN learned masks (ablation) | STS-DW-GCN · Human3.6M (1000 ms) | MPJPE (mm): 114.8 -> 115.7 | Hurt |
| sparsify adjacency via teacher-student framework (ablation) | STS-DW-GCN · Human3.6M (1000 ms) | MPJPE (mm): 114.8 -> 113.9 | Improved |
| teacher-student sparsification, parameter count (ablation) | STS-DW-GCN · Human3.6M (1000 ms) | parameters (K): 70.0 -> 58.6 | Improved |
| SeS-GCN vs DCT-RNN-GCN (state of the art) | SeS-GCN · Human3.6M (1000 ms) | MPJPE (mm): 113.9 vs 112.1 | No clear effect |
| SeS-GCN vs STS-GCN on CHICO short term | STS-GCN · CHICO (400 ms) | MPJPE (mm): 53.0 -> 48.8 | Improved |
| SeS-GCN vs STS-GCN on CHICO long term | STS-GCN · CHICO (1000 ms) | MPJPE (mm): 87.4 -> 85.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MSR-GCN — MSR-GCN: Multi-Scale Residual Graph Convolution Networks for Human Motion Prediction (ICCV 2021) · better on 4 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Space-Time-Separable Graph Convolutional Network for Pose Forecasting (ICCV 2021 · similarity 0.89)
- Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning (CVPR 2024 · similarity 0.86)
- FutureHuman3D: Forecasting Complex Long-Term 3D Human Behavior from Video Observations (CVPR 2024 · similarity 0.86)
- SimpliHuMoN: Simplifying Human Motion Prediction (NeurIPS 2026 · similarity 0.86)
- GCNext: Towards the Unity of Graph Convolutions for Human Motion Prediction (AAAI 2024 · similarity 0.85)
- World Motion Models: Flexible Sequence Modeling of SE(3) Trajectories (NeurIPS 2026 · 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.
