AI Paper Atlas

Research map / Time series / Long-term forecasting

Space-Time-Separable Graph Convolutional Network for Pose Forecasting

Theodoros Sofianos, Alessio Sampieri, Luca Franco, Fabio Galasso

ICCV 2021 · Time series › Long-term forecasting · approach: Spatiotemporal forecasting models · 211 citations (OpenAlex)

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 changedWhereMeasuredEffect
factor the space-time adjacency into separate spatial and temporal matrices within a single graph (vs separate GCNs for space and time) (ablation)STS-GCN · Human3.6M average, 80 msMPJPE (mm): 28.9 -> 10.1Improved
factor the space-time adjacency into separate spatial and temporal matrices within a single graph (vs separate GCNs for space and time) (ablation)STS-GCN · Human3.6M average, 1000 msMPJPE (mm): 79.9 -> 75.6Improved
factor the adjacency into A^s A^t instead of a full space-time matrix (ablation)STS-GCN · Human3.6M average, 80 msMPJPE (mm): 11.9 -> 10.1Improved
factor the adjacency into A^s A^t instead of a full space-time matrix (ablation)STS-GCN · Human3.6M average, 1000 msMPJPE (mm): 82.5 -> 75.6Improved
use layer-specific rather than shared adjacency matrices (ablation)STS-GCN · Human3.6M average, 80 msMPJPE (mm): 11.3 -> 10.1Improved
use layer-specific rather than shared adjacency matrices (ablation)STS-GCN · Human3.6M average, 1000 msMPJPE (mm): 76.9 -> 75.6Improved
use layer-specific rather than shared adjacency matrices (parameter cost) (ablation)STS-GCN · Human3.6M parametersparameters: 36.4k -> 57.5kHurt
use STS-GCN instead of DCT-RNN-GCNSTS-GCN · Human3.6M average, 1000 msMPJPE (mm): 112.1 -> 75.6Improved
use STS-GCN instead of DCT-RNN-GCNSTS-GCN · Human3.6M average, 80 msMPJPE (mm): 10.4 -> 10.1Improved
use STS-GCN instead of DCT-RNN-GCNSTS-GCN · 3DPW (trained on AMASS)MPJPE (mm): 73.7 -> 42.3Improved
use STS-GCN instead of DCT-RNN-GCNSTS-GCN · AMASS-BMLrub, 1000 msMPJPE (mm): 67.2 -> 45.5Improved
use STS-GCN instead of DCT-RNN-GCN in parametersSTS-GCN · Human3.6M parametersparameters: 3.4M -> 57.5kImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy