Research map / Video / Action recognition
Context-Aware Query Selection for Active Learning in Event Recognition
Mahmudul Hasan, Sujoy Paul, Anastasios I. Mourikis, Amit K. Roy–Chowdhury
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 context attributes (CAQS vs CAQS-NoC) (ablation) | CAQS (38% labels) · UCF50 | accuracy (%): 92.5 -> 98.2 | Improved |
| add context attributes (CAQS vs CAQS-NoC) (ablation) | CAQS (100% labels) · UCF50 | accuracy (%): 93.1 -> 99.1 | Improved |
| add context attributes (CAQS vs CAQS-NoC) (ablation) | CAQS (33% vs 47% labels) · VIRAT | accuracy (%): 75.9 -> 77.2 | Improved |
| add context attributes (CAQS vs CAQS-NoC) (ablation) | CAQS (100% labels) · VIRAT | accuracy (%): 76.3 -> 78.9 | Improved |
| add context attributes (CAQS vs CAQS-NoC) (ablation) | CAQS (42% labels) · MPII-Cooking | accuracy (%): 44.8 -> 49.4 | Improved |
| add context attributes (CAQS vs CAQS-NoC) (ablation) | CAQS (100% labels) · MPII-Cooking | accuracy (%): 45.2 -> 49.6 | Improved |
| use CAQS with 38% labels instead of CAAL greedy selection | CAAL (52% labels) · UCF50 | accuracy (%): 68.0 -> 98.2 | Improved |
| use CAQS with fewer labels instead of batch SSVM with all labels | SSVM · VIRAT | accuracy (%): 73.5 -> 77.2 | Improved |
| use CAQS instead of SCSG structure learning | SCSG · UCF50 | accuracy (%): 1.7% lower | Hurt |
| add weak teacher to strong teacher (ablation) | CAQS · UCF50/VIRAT/MPII-Cooking | accuracy | No clear effect |
| use weak teacher only (no manual labels after first batch) (ablation) | CAQS · UCF50/VIRAT/MPII-Cooking | accuracy | Hurt |
| add weak teacher on top of strong teacher (ablation) | CAQS · AVA | accuracy | Hurt |
Closest papers by meaning
- Non-Uniform Subset Selection for Active Learning in Structured Data (CVPR 2017 · similarity 0.88)
- Leaving Some Stones Unturned: Dynamic Feature Prioritization for Activity Detection in Streaming Video (ECCV 2016 · similarity 0.86)
- Contextual Active Model Selection (NeurIPS 2024 · similarity 0.86)
- Joint Prediction of Activity Labels and Starting Times in Untrimmed Videos (ICCV 2017 · similarity 0.85)
- HARDVS: Revisiting Human Activity Recognition with Dynamic Vision Sensors (AAAI 2024 · similarity 0.84)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 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.
