Research map / Video / Video segmentation and tracking
They Are Not Equally Reliable: Semantic Event Search Using Differentiated Concept Classifiers
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 |
|---|---|---|---|
| combine relevant, refined concept classifiers using warped spectral meta-learning | semantic event search · MEDTest 2014 | mAP: 12.13 -> 14.32 | Helped |
| combine relevant, refined concept classifiers using warped spectral meta-learning | semantic event search · MEDTest 2013 | mAP: 10.62 -> 13.46 | Helped |
| combine relevant, refined concept classifiers using warped spectral meta-learning | semantic event search · CCVsub | mAP: 22.52 -> 25.59 | Helped |
| add data-driven concept refinement to the discovered concepts (ablation) | warped spectral meta-learner · MEDTest 2014 | mAP: 12.52 -> 14.32 | Helped |
| add reliability-aware spectral weighting instead of Borda aggregation (ablation) | relevance-weighted, refined concept classifiers · MEDTest 2014 | mAP: 11.44 -> 14.32 | Helped |
| add relevance warping to the spectral meta-learner (ablation) | refined concept classifiers · MEDTest 2014 | mAP: 11.33 -> 14.32 | Helped |
| add relevance weighting and spectral reliability weighting (ablation) | refined concept classifiers · MEDTest 2014 | mAP: 9.12 -> 14.32 | Helped |
| combine wSML+ with an SVM trained on 10 positive examples | IDT features · MEDTest 2014 few-exemplar | mAP: 13.92 -> 16.98 | Helped |
| combine wSML+ with an SVM trained on 10 positive examples | IDT features · MEDTest 2013 few-exemplar | mAP: 18.08 -> 19.65 | Helped |
| use unsupervised accuracy estimation rather than estimating concept accuracy from labeled examples | spectral meta-learner · MEDTest 2014, up to roughly 8 labeled training videos | mAP | Helped |
Closest papers by meaning
- Dynamic Concept Composition for Zero-Example Event Detection (AAAI 2016 · similarity 0.90)
- Unified Embedding and Metric Learning for Zero-Exemplar Event Detection (CVPR 2017 · similarity 0.88)
- Beyond Grounding: Extracting Fine-Grained Event Hierarchies across Modalities (AAAI 2024 · similarity 0.86)
- MultiVENT 2.0: A Massive Multilingual Benchmark for Event-Centric Video Retrieval (CVPR 2025 · similarity 0.85)
- HyrCap: Hybrid Rank-Calibration of Action Proposals for Temporal Event Understanding (NeurIPS 2026 · similarity 0.84)
- Cross-modal Multi-task Learning for Multimedia Event Extraction (AAAI 2025 · similarity 0.84)
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.