Research map / Video / Action recognition
AOE-Net: Entities Interactions Modeling with Adaptive Attention Mechanism for Temporal Action Proposals Generation
Viet-Khoa Vo-Ho, Sang Truong, Kashu Yamazaki, Bhiksha Raj, Minh–Triet Tran, Ngan Le
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add objects beholder to actors+environment (soft attention) (ablation) | AOE-Net · THUMOS-14 TAPG (SNMS) | AR@50: 40.87 -> 42.60 | Improved |
| add AAM in place of soft attention on actors+environment (ablation) | AOE-Net · THUMOS-14 TAPG (SNMS) | AR@50: 40.87 -> 43.79 | Improved |
| add AAM in place of soft attention with objects (ablation) | AOE-Net · THUMOS-14 TAPG (SNMS) | AR@50: 42.60 -> 44.56 | Improved |
| add actors to environment features (ablation) | AOE-Net · THUMOS-14 TAPG (SNMS) | AR@50: 38.94 -> 40.87 | Improved |
| use only actors instead of environment (ablation) | AOE-Net · THUMOS-14 TAPG (SNMS) | AR@50: 38.94 -> 25.96 | Hurt |
| use only objects instead of environment (ablation) | AOE-Net · THUMOS-14 TAPG (SNMS) | AR@50: 38.94 -> 18.06 | Hurt |
| use AAM instead of hard attention (ablation) | AOE-Net · THUMOS-14 TAPG (SNMS) | AR@50: 43.74 -> 44.56 | Improved |
| use AAM instead of soft attention (ablation) | AOE-Net · THUMOS-14 TAPG (SNMS) | AR@50: 42.60 -> 44.56 | Improved |
| use AAM instead of hard attention (ablation) | AOE-Net · ActivityNet-1.3 TAPG | AR@100: 77.11 -> 77.67 | Improved |
| use AAM instead of soft attention (ablation) | AOE-Net · ActivityNet-1.3 TAPG | AUC (test): 69.23 -> 70.10 | Improved |
| add objects beholder with K=20 objects (versus K=0) (ablation) | AOE-Net · ActivityNet-1.3 TAPG | AR@100: 77.02 -> 77.67 | Improved |
| increase number of objects K beyond 40 to 50 (ablation) | AOE-Net · ActivityNet-1.3 TAPG | AR@100: 77.67 -> 77.24 | Hurt |
Closest papers by meaning
- TURN TAP: Temporal Unit Regression Network for Temporal Action Proposals (ICCV 2017 · similarity 0.89)
- Temporal Action Proposal Generation with Background Constraint (AAAI 2022 · similarity 0.88)
- About Time: Advances, Challenges, and Outlooks of Action Understanding (IJCV 2025 · similarity 0.88)
- AVA: A Video Dataset of Spatio-Temporally Localized Atomic Visual Actions (CVPR 2018 · similarity 0.88)
- EAN: Event Adaptive Network for Enhanced Action Recognition (IJCV 2022 · similarity 0.88)
- Accurate Temporal Action Proposal Generation with Relation-Aware Pyramid Network (AAAI 2020 · similarity 0.88)
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.
