Research map / Video / Action recognition
Boosting Weakly-Supervised Temporal Action Localization With Text Information
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 |
|---|---|---|---|
| Replace the convolution classifier with text-segment mining (ablation) | WTAL on THUMOS14 · THUMOS14 | Avg mAP: 36.3 -> 37.2 | Helped |
| Add VLC reconstruction and consistency losses (ablation) | Video-only baseline WTAL · THUMOS14 | Avg mAP: 34.9 -> 35.9 | Helped |
| Add VLC contrastive loss L_c (ablation) | Baseline + VLC · THUMOS14 | Avg mAP: 35.9 -> 36.3 | Helped |
| Use a learnable prompt rather than a handcrafted prompt in TSM (ablation) | TSM · THUMOS14 | Avg mAP: 36.1 -> 37.2 | Helped |
| Use MSE consistency instead of no constraint, shared attention, KL, or MAE (ablation) | TSM and VLC · THUMOS14 | Avg mAP: 35.4 -> 37.2; 35.8 -> 37.2; 36.9 -> 37.2; 36.4 -> 37.2 | Helped |
| Use Transformer rather than GRU as VLC language reconstructor (ablation) | VLC · THUMOS14 | Avg mAP: 36.7 -> 37.2 | Helped |
| Use Transformer rather than LSTM as VLC language reconstructor (ablation) | VLC · THUMOS14 | Avg mAP: 36.1 -> 37.2 | Helped |
| Use completion prompt “a video of the [CLS]” rather than “a [CLS]” (ablation) | VLC · THUMOS14 | Avg mAP: 36.8 -> 37.2 | Helped |
| Use completion prompt “a video of the [CLS]” rather than “a video of action [CLS]” (ablation) | VLC · THUMOS14 | Avg mAP: 36.6 -> 37.2 | Helped |
| Add TSM to BaSNet (ablation) | BaSNet · THUMOS14 | Avg mAP: 27.3 -> 29.5 | Helped |
| Add VLC to BaSNet (ablation) | BaSNet · THUMOS14 | Avg mAP: 27.3 -> 30.2 | Helped |
| Add TSM and VLC to BaSNet (ablation) | BaSNet · THUMOS14 | Avg mAP: 27.3 -> 30.6 | Helped |
Closest papers by meaning
- Weakly Supervised Temporal Action Localization via Representative Snippet Knowledge Propagation (CVPR 2022 · similarity 0.92)
- Weakly-Supervised Temporal Action Localization by Inferring Salient Snippet-Feature (AAAI 2024 · similarity 0.89)
- Bridge the Gap: From Weak to Full Supervision for Temporal Action Localization with PseudoFormer (CVPR 2025 · similarity 0.89)
- Weakly Supervised Temporal Action Localization via Dual-Prior Collaborative Learning Guided by Multimodal Large Language Models (CVPR 2025 · similarity 0.89)
- HR-Pro: Point-Supervised Temporal Action Localization via Hierarchical Reliability Propagation (AAAI 2024 · similarity 0.89)
- Similar Modality Enhancement and Action Consistency Learning for Weakly Supervised Temporal Action Localization (AAAI 2025 · similarity 0.88)
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.