AI Paper Atlas

Research map / 3D vision / Human pose estimation

Trajectory-Aware Body Interaction Transformer for Multi-Person Pose Forecasting

CVPR 2023 · 3D vision › Human pose estimation · approach: Human motion prediction · 30 citations (OpenAlex)

Read the paperCheck 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 changedWhereMeasuredEffect
Add Temporal Body Partition Module (TBPM) (ablation)TBIFormer · CMU-Mocap (UMPM)JPE 1.0s: 195 -> 182Helped
Add TRPE (trajectory-aware relative position encoding) over Euclidean encoding (ablation)TBIFormer · CMU-Mocap (UMPM)JPE 1.0s: 191 -> 182Helped
Add TRPE (ablation)TBIFormer · CMU-Mocap (UMPM)JPE 1.0s: 186 -> 182Helped
Add person identity encoding (IE) (ablation)TBIFormer · CMU-Mocap (UMPM)JPE 1.0s: 188 -> 186 (w/o IE,TRPE vs w/o TRPE)No clear effect
Replace SBI-MSA with standard self-attention (ablation)TBIFormer · CMU-Mocap (UMPM)JPE 1.0s: 208 -> 182 (adding SBI-MSA)Helped
Use TBIFormer trained on CMU-Mocap with motion-based TRPE vs prior baselines (random person permutation test)TBIFormer · CMU-Mocap (UMPM) permutationJPE 1.0s: 182 -> 184 under permutationNo clear effect
Proposed TBIFormer vs multi-person baseline MRT on out-of-domain datasetsTBIFormer · MuPoTS-3D, Mix1, Mix2JPE overall: MuPoTS 217->195; Mix1 124->121; Mix2 126->117Helped

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