Research map / Video / Action recognition
Colar: Effective and Efficient Online Action Detection by Consulting Exemplars
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 category exemplars and the static branch to the dynamic branch (ablation) | Colar · THUMOS14 | mAP: 65.2 -> 65.8 | Helped |
| Add category exemplars and the static branch to the dynamic branch (ablation) | Colar · TVSeries | mAP: 86.3 -> 86.9 | Helped |
| Add category exemplars and the static branch to the dynamic branch (ablation) | Colar · HDD | mAP: 29.5 -> 29.9 | Helped |
| Add temporal modeling with the dynamic branch to the static branch (ablation) | Colar · THUMOS14 | mAP: 58.1 -> 65.8 | Helped |
| Add temporal modeling with the dynamic branch to the static branch (ablation) | Colar · TVSeries | mAP: 83.5 -> 86.9 | Helped |
| Add temporal modeling with the dynamic branch to the static branch (ablation) | Colar · HDD | mAP: 26.4 -> 29.9 | Helped |
| Add consistency loss between dynamic and static predictions (ablation) | Colar · THUMOS14 | mAP: 65.8 -> 66.9 | Helped |
| Add consistency loss between dynamic and static predictions (ablation) | Colar · TVSeries | mAP: 86.9 -> 88.1 | Helped |
| Add consistency loss between dynamic and static predictions (ablation) | Colar · HDD | mAP: 29.9 -> 30.6 | Helped |
| Use end-to-end backbone training instead of extracting fixed-backbone features | Colar · THUMOS14 | mAP: 53.4 -> 58.8 | Helped |
| Use RGB features only instead of RGB and flow features | Colar · THUMOS14 | mAP: 66.9 -> 52.1 | Hurt |
| Use flow features only instead of RGB and flow features | Colar · THUMOS14 | mAP: 66.9 -> 59.6 | Hurt |
Closest papers by meaning
- Context-Enhanced Memory-Refined Transformer for Online Action Detection (CVPR 2025 · similarity 0.84)
- OnlineTAS: An Online Baseline for Temporal Action Segmentation (NeurIPS 2024 · similarity 0.84)
- Stable Mean Teacher for Semi-supervised Video Action Detection (AAAI 2025 · similarity 0.84)
- Does Video-Text Pretraining Help Open-Vocabulary Online Action Detection? (NeurIPS 2024 · similarity 0.84)
- Benchmarking the Robustness of Temporal Action Detection Models Against Temporal Corruptions (CVPR 2024 · similarity 0.83)
- Dual DETRs for Multi-Label Temporal Action Detection (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.