AI Paper Atlas

Research map / Video / Action recognition

Learning Asynchronous and Sparse Human-Object Interaction in Videos

Romero Morais, Vuong Le, Svetha Venkatesh, Truyen Tran

CVPR 2021 · Video › Action recognition · approach: Disentangled action recognition · 45 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
Jointly discover segment boundaries and labels with sparse, asynchronous interaction modelingASSIGN · CAD-120, joint segmentationF1@0.10: Relational BiRNN 79.2 ± 2.5 -> 88.0 ± 1.8; Independent BiRNN 70.2 ± 5.5 -> 88.0 ± 1.8Improved
Use ASSIGN for multi-entity interactions and segment-level decisionsBimanual Actions models · Bimanual Actions, joint segmentationF1@0.10: Relational BiRNN 77.7 ± 3.9 -> 84.0 ± 2.0Improved
Add spatial message passing (ablation)ASSIGN · CAD-120, joint segmentation; sub-activityF1@0.10: 74.5 -> 88.0Improved
Add spatial message passing (ablation)ASSIGN · CAD-120, joint segmentation; object affordanceF1@0.50: 74.4 -> 82.4Improved
Add segmentation loss for joint segmentation and labeling (ablation)ASSIGN · CAD-120, joint segmentation; sub-activityF1@0.10: 84.3 -> 88.0Improved
Use sparse asynchronous rather than dense updates (ablation)ASSIGN · CAD-120, joint segmentation; sub-activityF1@0.50: 70.3 -> 73.8Improved
Use sparse asynchronous rather than dense updates (ablation)ASSIGN · CAD-120, joint segmentation; object affordanceF1@0.50: 79.8 -> 82.4Improved
Add dense-model pre-training (ablation)ASSIGN · CAD-120, joint segmentation; sub-activityF1@0.50: 71.6 -> 73.8Improved
Add dense-model pre-training (ablation)ASSIGN · CAD-120, joint segmentation; object affordanceF1@0.50: 78.9 -> 82.4Improved
Add anticipation loss (ablation)ASSIGN · CAD-120, joint segmentation; sub-activityF1@0.10: 86.2 -> 88.0Improved
Add anticipation loss (ablation)ASSIGN · CAD-120, joint segmentation; object affordanceF1@0.50: 80.4 -> 82.4Improved

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