Research map / Video / Video segmentation and tracking
Dynamic Concept Composition for Zero-Example Event Detection
Read the paperCheck 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 |
|---|---|---|---|
| learn per-video dynamic concept weights with infinity push (DCC) instead of PCF | PCF · MEDTest 2014 | mAP (%): 11.44 -> 13.37 | Helped |
| learn per-video dynamic concept weights (DCC) instead of PCF | PCF · MEDTest 2013 | mAP (%): 9.96 -> 12.64 | Helped |
| learn per-video dynamic concept weights (DCC) instead of PCF | PCF · CCV-sub | mAP (%): 23.87 -> 24.36 | Helped |
| select informative concepts instead of using all primitive concepts | Prim · MEDTest 2014 | mAP (%): 6.40 -> 9.55 | Helped |
| select informative concepts instead of using all concepts | Prim · MEDTest 2013 | mAP (%): 7.07 -> 7.94 | Helped |
| weight selected concepts differently (Fu) instead of equal Borda aggregation | Bor · MEDTest 2014 | mAP (%): 10.21 -> 11.05 | Helped |
| combine DCC (0Ex) with IDT (10Ex) supervised classifier | IDT (10Ex) · MEDTest 2014 | mAP (%): 13.92 -> 16.37 | Mixed |
| combine DCC (0Ex) with IDT (10Ex) supervised classifier | IDT (10Ex) · MEDTest 2013 | mAP (%): 18.08 -> 19.24 | Helped |
| DCC with no labeled examples versus supervised IDT with 10 positives | IDT (10Ex) · MEDTest 2014 | mAP (%): 13.37 vs 13.92 | No clear effect |
Closest papers by meaning
- Text Prompt with Normality Guidance for Weakly Supervised Video Anomaly Detection (CVPR 2024 · similarity 0.83)
- Learning Event Completeness for Weakly Supervised Video Anomaly Detection (ICML 2025 · similarity 0.82)
- A Unified Reasoning Framework for Holistic Zero-Shot Video Anomaly Analysis (NeurIPS 2025 · similarity 0.82)
- Text-Infused Attention and Foreground-Aware Modeling for Zero-Shot Temporal Action Detection (NeurIPS 2024 · similarity 0.82)
- Beyond Grounding: Extracting Fine-Grained Event Hierarchies across Modalities (AAAI 2024 · similarity 0.82)
- Prompt-Enhanced Multiple Instance Learning for Weakly Supervised Video Anomaly Detection (CVPR 2024 · similarity 0.81)
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.