Research map / Video / Action recognition
Directional Temporal Modeling for Action Recognition
Xinyu Li, Bing Shuai, Joseph Tighe
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 single-scale CIDC to I3D-50 (ablation) | I3D-50 · UCF101 | top-1 accuracy: 92.9 -> 95.2 | Improved |
| add single-scale CIDC to I3D-50 (ablation) | I3D-50 · HMDB51 | top-1 accuracy: 69.1 -> 73.6 | Improved |
| add single-scale CIDC to I3D-50 (ablation) | I3D-50 · Kinetics-400 | top-1 accuracy: 73.8 -> 74.0 | Improved |
| make CIDC multi-scale instead of single-scale (ablation) | CIDC network on I3D-50 · UCF101 | top-1 accuracy: 95.2 -> 95.9 | Improved |
| make CIDC multi-scale instead of single-scale (ablation) | CIDC network on I3D-50 · HMDB51 | top-1 accuracy: 73.6 -> 74.1 | Improved |
| make CIDC multi-scale instead of single-scale (ablation) | CIDC network on I3D-50 · Kinetics-400 | top-1 accuracy: 74.0 -> 74.4 | Improved |
| add spatial attention propagation to multi-scale CIDC (ablation) | CIDC network on I3D-50 · UCF101 | top-1 accuracy: 95.9 -> 97.2 | Improved |
| add spatial attention propagation to multi-scale CIDC (ablation) | CIDC network on I3D-50 · HMDB51 | top-1 accuracy: 74.1 -> 74.9 | Improved |
| add spatial attention propagation to multi-scale CIDC (ablation) | CIDC network on I3D-50 · Kinetics-400 | top-1 accuracy: 74.4 -> 74.5 | No clear effect |
| make CIDC bi-directional instead of non-directional (ablation) | CIDC network on I3D-50 · UCF101 | top-1 accuracy: 94.1 -> 97.2 | Improved |
| make CIDC bi-directional instead of non-directional (ablation) | CIDC network on I3D-50 · HMDB51 | top-1 accuracy: 72.5 -> 74.9 | Improved |
| make CIDC bi-directional instead of uni-directional (ablation) | CIDC network on I3D-50 · Kinetics-400 | top-1 accuracy: 74.2 -> 74.5 | Improved |
Closest papers by meaning
- T-C3D: Temporal Convolutional 3D Network for Real-Time Action Recognition (AAAI 2018 · similarity 0.87)
- Two-Stream Region Convolutional 3D Network for Temporal Activity Detection (TPAMI 2019 · similarity 0.87)
- R-C3D: Region Convolutional 3D Network for Temporal Activity Detection (ICCV 2017 · similarity 0.86)
- ACTION-Net: Multipath Excitation for Action Recognition (CVPR 2021 · similarity 0.86)
- Temporal Convolutional Networks for Action Segmentation and Detection (CVPR 2017 · similarity 0.85)
- Spatio-Temporal Channel Correlation Networks for Action Classification (ECCV 2018 · similarity 0.85)
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.
