Research map / Video / Action recognition
Dual-AI: Dual-Path Actor Interaction Learning for Group Activity Recognition
Check 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 changed | Where | Measured | Effect |
|---|---|---|---|
| fuse the spatial-temporal and temporal-spatial interaction paths (ablation) | Dual-AI · Limited Volleyball | accuracy: 89.3, 89.5 -> 90.8 | Helped |
| fuse the spatial-temporal and temporal-spatial interaction paths (ablation) | Dual-AI · Weak Volleyball-M | accuracy: 93.0, 92.6 -> 94.2 | Helped |
| fuse the spatial-temporal and temporal-spatial interaction paths (ablation) | Dual-AI · Full Volleyball | accuracy: 92.2, 92.1 -> 93.3 | Helped |
| add frame-frame actor contrastive loss (ablation) | Dual-AI · Volleyball 50% data | accuracy: 90.8 -> 91.2 | Helped |
| add frame-video actor contrastive loss (ablation) | Dual-AI · Volleyball 50% data | accuracy: 90.8 -> 91.0 | Helped |
| add video-video actor contrastive loss (ablation) | Dual-AI · Volleyball 50% data | accuracy: 90.8 -> 91.6 | Helped |
| add all MAC-loss components (ablation) | Dual-AI · Volleyball 50% data | accuracy: 90.8 -> 92.1 | Helped |
| add all MAC-loss components (ablation) | Dual-AI · Volleyball 100% data | accuracy: 93.3 -> 94.0 | Helped |
| use late scene-context fusion instead of no scene fusion (ablation) | Dual-AI · Volleyball 50% data | accuracy: 92.1 -> 92.7 | Helped |
| use late scene-context fusion instead of no scene fusion (ablation) | Dual-AI · Volleyball 100% data | accuracy: 94.0 -> 94.4 | Helped |
| add optical flow input | Dual-AI · Volleyball | group activity accuracy: 94.4 -> 95.4 | Helped |
| fuse RGB and optical flow | Dual-AI · NBA | accuracy: 51.5, 56.8 -> 58.1 | Helped |
Closest papers by meaning
- Learning Actor Relation Graphs for Group Activity Recognition (CVPR 2019 · similarity 0.90)
- Actor-Transformers for Group Activity Recognition (CVPR 2020 · similarity 0.90)
- Detector-Free Weakly Supervised Group Activity Recognition (CVPR 2022 · similarity 0.87)
- Progressive Relation Learning for Group Activity Recognition (CVPR 2020 · similarity 0.87)
- Convolutional Relational Machine for Group Activity Recognition (CVPR 2019 · similarity 0.87)
- Learning from Synthetic Human Group Activities (CVPR 2024 · similarity 0.83)
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.