AI Paper Atlas

Research map / Video / Action recognition

Actor-Context-Actor Relation Network for Spatio-Temporal Action Localization

Junting Pan, Siyu Chen, Mike Zheng Shou, Yu Liu, Jing Shao, Hongsheng Li

CVPR 2021 · Video › Action recognition · approach: Weakly supervised temporal action localization · 144 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
Add high-order actor-context-actor reasoning (HR²O) (ablation)SlowFast R-50 baseline · AVA v2.2mAP: 25.39 -> 27.83Improved
Add Actor-Context Feature Bank (ablation)HR²O · AVA v2.2mAP: 27.83 -> 28.84Improved
Replace ACFB with Long-term Feature Bank (ablation)HR²O · AVA v2.2mAP: 27.83 -> 27.75Hurt
Use modified non-local attention rather than average aggregation (ablation)HR²O · AVA v2.2mAP: 26.97 -> 27.83Improved
Use modified non-local attention rather than a relation network (ablation)HR²O · AVA v2.2mAP: 27.18 -> 27.83Improved
Encode actor-context relations before actor-actor reasoning (ablation)ACAR relation reasoning · AVA v2.2mAP: 27.62 -> 27.83Improved
Increase HR²O depth from one to two non-local blocks (ablation)HR²O · AVA v2.2mAP: 27.63 -> 27.83Improved
Increase HR²O depth from two to three non-local blocks (ablation)HR²O · AVA v2.2mAP: 27.83 -> 27.25Hurt
Add actor-context-actor relation modeling instead of only first-order relation methods (ablation)SlowFast R-50 baseline · AVA v2.2mAP: 26.10 -> 27.83Improved
Use ACAR without ACFB rather than the SlowFast baselineSlowFast R-50 · UCF101-24 split 1frame-mAP: 82.4 -> 84.3Improved
Add ACAR modelingR-101 SlowFast model · AVA-Kinetics valmAP: +2.86 mAPImproved
Use ACAR-Net rather than the SlowFast R-101 baselineR-101 SlowFast model · AVA-Kinetics valmAP: 32.98 -> 35.84Improved

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