AI Paper Atlas

Research map / 3D vision / Human pose estimation

On the Benefits of 3D Pose and Tracking for Human Action Recognition

Jathushan Rajasegaran, Georgios Pavlakos, Angjoo Kanazawa, Christoph Feichtenhofer, Jitendra Malik

CVPR 2023 · 3D vision › Human pose estimation · approach: Physics and sensor-based motion capture · 49 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
Add person tracking to MViT appearance features (ablation)MViT action recognition on AVA · AVAmAP: 40.2 -> 41.4 (+1.2)Improved
Add 3D pose features to tracked MViT features (ablation)MViT + Tracking · AVAmAP: 41.4 -> 42.3 (+0.9)Improved
Add 3D pose and tracking to appearance featuresMViT on AVA v2.2 · AVA v2.2mAP: +2.8 mAPImproved
Add LART track modeling to Hiera appearance featuresHiera on AVA v2.2 · AVA v2.2mAP: +2.5 mAPImproved
Model other people's 3D tracks alongside the person-of-interest (ablation)LART-pose · AVAPerson Interactions mAP: +2.1 mAPImproved
Model five people's 3D tracks instead of only the target person's track (ablation)LART-pose · AVAmAP: 22.9 -> 24.1; 2.0 mAPImproved
Increase the number of people in the pose-model context (ablation)LART-pose · AVAmAPImproved
Use explicit 3D pose and tracking instead of appearance-only MViT featuresLART-pose versus JMRN · AVAmAP: +10.0 mAPImproved

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