Research map / Architectures and efficiency / RNNs and normalization
Recurrent Modeling of Interaction Context for Collective 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 |
|---|---|---|---|
| Use hierarchical recurrent interaction-context modeling instead of pooling-SVM | Collective activity recognition · Collective Activity Dataset | MPCA (%): 85.8 -> 89.4 | Helped |
| Use hierarchical recurrent interaction-context modeling | Collective activity recognition · Collective Activity Dataset | MPCA (%): 90.8 -> 89.4 | Hurt |
| Use recurrent interaction-context modeling with motion threshold µ = 0 | Group activity recognition · Choi’s New Dataset | MCA (%): 83.0 -> 89.4 | Helped |
| Use recurrent interaction-context modeling with motion threshold µ = 1 | Group activity recognition · Choi’s New Dataset | MPCA (%): 80.3 -> 85.2 | Helped |
| Set motion threshold µ to 1 pixel instead of 0 (ablation) | Recurrent interaction-context model · Choi’s New Dataset | MCA (%): 89.4 -> 89.2 | Hurt |
| Set motion threshold µ to 1 pixel instead of 0 (ablation) | Recurrent interaction-context model · Choi’s New Dataset | MPCA (%): 84.6 -> 85.2 | Helped |
| Set motion threshold µ to 2 pixels instead of 0 (ablation) | Recurrent interaction-context model · Choi’s New Dataset | MCA (%): 89.4 -> 87.3 | Hurt |
| Set motion threshold µ to 2 pixels instead of 0 (ablation) | Recurrent interaction-context model · Choi’s New Dataset | MPCA (%): 84.6 -> 84.5 | No clear effect |
Closest papers by meaning
- A Hierarchical Deep Temporal Model for Group Activity Recognition (CVPR 2016 · similarity 0.91)
- Learning Visual Context for Group Activity Recognition (AAAI 2021 · similarity 0.87)
- Detector-Free Weakly Supervised Group Activity Recognition (CVPR 2022 · similarity 0.86)
- Structure Inference Machines: Recurrent Neural Networks for Analyzing Relations in Group Activity Recognition (CVPR 2016 · similarity 0.86)
- Learning Group Activity Features Through Person Attribute Prediction (CVPR 2024 · similarity 0.83)
- Learning Human-Object Interaction as Groups (NeurIPS 2025 · 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.