AI Paper Atlas

Research map / 3D vision / Human pose estimation

Forecasting Characteristic 3D Poses of Human Actions

Christian Diller, Thomas Funkhouser, Angela Dai

CVPR 2022 · 3D vision › Human pose estimation · approach: Human motion prediction · 22 citations (OpenAlex)

Check 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
Use probabilistic volumetric heatmap predictions instead of a deterministic MLP head (ablation)Characteristic 3D Pose Forecasting · GRABMPJPE: 0.064 -> 0.054Improved
Use probabilistic volumetric heatmap predictions instead of a deterministic MLP head (ablation)Characteristic 3D Pose Forecasting · Human3.6MMPJPE: 0.108 -> 0.092Improved
Sample joints autoregressively rather than independently (ablation)Characteristic 3D Pose Forecasting · GRABMPJPE: 0.077 -> 0.054Improved
Add per-voxel offset prediction (ablation)Characteristic 3D Pose Forecasting · GRABMPJPE: 0.132 -> 0.054Improved
Add pose refinement (ablation)Characteristic 3D Pose Forecasting · GRABMPJPE: 0.057 -> 0.054Improved
Increase pose samples from the default to k=50 (ablation)Characteristic 3D Pose Forecasting · GRABMPJPE: 0.054 -> 0.049Improved
Use discretized cross-entropy heatmap loss rather than ℓ1 regression loss (ablation)Characteristic 3D Pose Forecasting · Human3.6MMPJPE: 0.198 -> 0.092Improved
Use discretized cross-entropy heatmap loss rather than ℓ2 regression loss (ablation)Characteristic 3D Pose Forecasting · GRABMPJPE: 0.130 -> 0.054Improved
Use right hand → left hand → rest autoregressive ordering (ablation)Characteristic 3D Pose Forecasting · GRABDiversity: 0.049 -> 0.105Improved
Use a 16³ grid with offset prediction rather than a 32³ grid with offsets (ablation)Characteristic 3D Pose Forecasting · GRABMPJPE: 0.066 -> 0.054Improved
Provide ground-truth action labels as additional input (ablation)Characteristic 3D Pose Forecasting · GRABIS: 4.153 { $\pm 0.87$} -> 1.085 { $\pm 0.02$}Hurt
Use the proposed characteristic-pose method rather than DLowCharacteristic 3D Pose Forecasting · GRABMPJPE: 0.071 -> 0.054Improved

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