Research map / Video / Action recognition
Hybrid Active Learning via Deep Clustering for Video Action 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 |
|---|---|---|---|
| Use hybrid CLAUS video-and-frame selection instead of the listed active-learning baselines | Video action detection · UCF-101-24, 1% annotations | v-mAP | Helped |
| Use hybrid CLAUS video-and-frame selection instead of the listed active-learning baselines | Video action detection · J-HMDB-21, 1% annotations | v-mAP | Helped |
| Add cluster-based diversity to video selection (ablation) | CLAUS · UCF-101-24 and J-HMDB-21 | v-mAP | Helped |
| Add STeW weighting to interpolated pseudo-label training (ablation) | Video action detection with CLAUS · UCF-101-24 | v-mAP | Helped |
| Use interpolated pseudo-labels rather than loss only on annotated frames (ablation) | Video action detection with CLAUS · UCF-101-24 | v-mAP | Helped |
| Use CLAUS scoring rather than random, equidistant, entropy, or uncertainty scoring (ablation) | Active-learning selection with STeW loss · UCF-101-24 and J-HMDB-21 | v-mAP | Helped |
| Increase both selected videos and annotated frames rather than increasing frames alone (ablation) | Active-learning annotation schedule · UCF-101-24 | v-mAP | Helped |
| Select videos and frames with hybrid selection instead of inter-sample-only selection (ablation) | Active learning for video action detection · UCF-101-24 and J-HMDB-21 | v-mAP | Helped |
| Select videos and frames with hybrid selection instead of intra-sample-only selection (ablation) | Active learning for video action detection · UCF-101-24 and J-HMDB-21 | v-mAP | Helped |
| Use CLAUS instead of random selection | Video action detection · UCF-101-24 | v-mAP: 72.2 vs 73.6 | Helped |
| Use CLAUS instead of random selection | Video action detection · J-HMDB-21 | v-mAP: 71.5 vs 73.1 | Helped |
Closest papers by meaning
- Are all Frames Equal? Active Sparse Labeling for Video Action Detection (NeurIPS 2022 · similarity 0.90)
- Semi-supervised Active Learning for Video Action Detection (AAAI 2024 · similarity 0.89)
- Uncertainty Guided Collaborative Training for Weakly Supervised Temporal Action Detection (CVPR 2021 · similarity 0.87)
- A flexible model for training action localization with varying levels of supervision (NeurIPS 2018 · similarity 0.86)
- Stable Mean Teacher for Semi-supervised Video Action Detection (AAAI 2025 · similarity 0.85)
- Rethinking Pseudo-Label Guided Learning for Weakly Supervised Temporal Action Localization from the Perspective of Noise Correction (AAAI 2025 · 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.