Research map / Video / Action recognition
Equivalent Classification Mapping for Weakly Supervised Temporal Action Localization
Tao Zhao, Junwei Han, Le Yang, Dingwen Zhang
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 |
|---|---|---|---|
| share classifier between pre- and post-classification streams (ablation) | ECM · THUMOS14 | mAP@0.5: 19.6 -> 22.8 | Improved |
| add aggregation-to-classification consistency loss (ablation) | ECM · THUMOS14 | mAP@0.5: 22.8 -> 24.2 | Improved |
| add classification-to-classification consistency loss (ablation) | ECM · THUMOS14 | mAP@0.5: 24.2 -> 29.1 | Improved |
| share classifiers between streams in full model (ablation) | ECM · THUMOS14 | mAP@0.5: 23.4 -> 29.1 | Improved |
| learn both streams jointly instead of merging independent results (ablation) | ECM · THUMOS14 | mAP@0.5: 21.4 -> 29.1 | Improved |
| learn jointly instead of weighted CAS fusion (ablation) | ECM · THUMOS14 | mAP@0.5: 20.8 -> 29.1 | Improved |
| equivalent mapping vs pre-classification only (ablation) | ECM · THUMOS14 | frame accuracy: 42.7% -> 54.3% | Improved |
| equivalent mapping vs post-classification only (ablation) | ECM · THUMOS14 | frame accuracy: 39.0% -> 54.3% | Improved |
| ECM vs BaSNet | ECM · THUMOS14 | mAP@0.5: 27.0 -> 29.1 | Improved |
| ECM vs DGAM | ECM · THUMOS14 | mAP@0.5: 28.8 -> 29.1 | Improved |
| ECM vs BaSNet | ECM · ActivityNet v1.2 | avg mAP: 24.3 -> 25.5 | Improved |
| ECM vs ACL | ECM · ActivityNet v1.3 | avg mAP: 22.5 -> 23.5 | Improved |
Closest papers by meaning
- Two-Stream Networks for Weakly-Supervised Temporal Action Localization With Semantic-Aware Mechanisms (CVPR 2023 · similarity 0.91)
- Two-Branch Relational Prototypical Network for Weakly Supervised Temporal Action Localization (TPAMI 2021 · similarity 0.91)
- Cross-Task Relation-Aware Consistency for Weakly Supervised Temporal Action Detection (TPAMI 2025 · similarity 0.90)
- Distilling Vision-Language Pre-Training To Collaborate With Weakly-Supervised Temporal Action Localization (CVPR 2023 · similarity 0.90)
- Weakly-Supervised Temporal Action Localization by Inferring Salient Snippet-Feature (AAAI 2024 · similarity 0.90)
- Similar Modality Enhancement and Action Consistency Learning for Weakly Supervised Temporal Action Localization (AAAI 2025 · similarity 0.89)
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.
