Research map / Video / Action recognition
A Structured Model for Action Detection
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 |
|---|---|---|---|
| add actor tubelets and graph convolutions to aggregate features across an actor's track (ablation) | I3D action detector · AVA validation | mAP: 16.7 -> 21.1 | Helped |
| add a person similarity graph over ROIs (ablation) | I3D action detector · AVA validation | mAP: 16.7 -> 20.1 | Helped |
| add an object similarity graph over ROIs (ablation) | I3D action detector · AVA validation | mAP: 16.7 -> 20.3 | Helped |
| add a hard relation graph to actor tubelets (ablation) | I3D action detector · AVA validation | mAP: 21.1 -> 21.5 | Helped |
| add a soft relation graph to actor tubelets (ablation) | I3D action detector · AVA validation | mAP: 21.1 -> 22.2 | Helped |
| replace soft relation aggregation with hard relation selection (ablation) | Actor tubelets relation model · AVA validation | mAP: 22.2 -> 21.5 | Hurt |
| add actor tubelets instead of using the baseline temporal average (ablation) | I3D action detector · AVA validation, human pose | mAP: 35.7 -> 40.6 | Helped |
| add actor tubelets instead of using the baseline temporal average (ablation) | I3D action detector · AVA validation, object manipulation | mAP: 8.9 -> 13.4 | Helped |
| add actor tubelets instead of using the baseline temporal average (ablation) | I3D action detector · AVA validation, human interaction | mAP: 16.9 -> 20.9 | Helped |
| add the soft relation graph to actor tubelets (ablation) | I3D action detector · AVA validation, human pose | mAP: 40.6 -> 41.9 | Helped |
| add the soft relation graph to actor tubelets (ablation) | I3D action detector · AVA validation, object manipulation | mAP: 13.4 -> 14.3 | Helped |
| add the soft relation graph to actor tubelets (ablation) | I3D action detector · AVA validation, human interaction | mAP: 20.9 -> 22.0 | Helped |
Closest papers by meaning
- Graph Attention Based Proposal 3D ConvNets for Action Detection (AAAI 2020 · similarity 0.89)
- VideoNet: A Large-Scale Dataset for Domain-Specific Action Recognition (CVPR 2026 · similarity 0.88)
- T-C3D: Temporal Convolutional 3D Network for Real-Time Action Recognition (AAAI 2018 · similarity 0.88)
- VideoCapsuleNet: A Simplified Network for Action Detection (NeurIPS 2018 · similarity 0.87)
- Deep Analysis of CNN-Based Spatio-Temporal Representations for Action Recognition (CVPR 2021 · similarity 0.87)
- DVANet: Disentangling View and Action Features for Multi-View Action Recognition (AAAI 2024 · similarity 0.86)
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.