Research map / Video / Action recognition
Distilling Vision-Language Pre-Training To Collaborate With Weakly-Supervised Temporal Action Localization
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 |
|---|---|---|---|
| distill complementary foreground and background pseudo-labels with alternating B and F steps (ablation) | CBP–VLP collaboration · THUMOS14 | mAP@IoU 0.3-0.7 average: 37.8 -> 40.8 | Helped |
| distill foreground pseudo-labels from vanilla VLP only (ablation) | CBP–VLP collaboration · THUMOS14 | mAP@IoU 0.3-0.7 average: 27.7 -> 40.8 | Hurt |
| add confident foreground-background knowledge distillation (ablation) | CBP–VLP collaboration · THUMOS14 | mAP@IoU 0.3-0.7 average: 30.9 -> 40.8 | Helped |
| add foreground-background contrastive loss (ablation) | CBP–VLP collaboration · THUMOS14 | mAP@IoU 0.3-0.7 average: 36.0 -> 40.8 | Helped |
| use the distillation-collaboration framework rather than DELU | WTAL · THUMOS14 | mAP@IoU 0.3-0.7 average: 37.4 -> 40.8 | Helped |
| use the distillation-collaboration framework rather than DELU | WTAL · ActivityNet1.2 | average mAP: 26.9 -> 29.6 | Helped |
| add CLIP features by averaging or concatenating them with I3D features | CO2-Net and DELU · THUMOS14 | mAP@IoU 0.3-0.7 average: CO2-Net: 35.7 -> 32.2; DELU: 37.4 -> 35.8 | Mixed |
| apply the framework using UM as the CBP branch (ablation) | UM · THUMOS14 | mAP@IoU 0.3: 49.0 -> 53.9 | Helped |
| apply the framework using CO2-Net as the CBP branch (ablation) | CO2-Net · THUMOS14 | mAP@IoU 0.3: 54.5 -> 56.2 | Helped |
Closest papers by meaning
- Bridge the Gap: From Weak to Full Supervision for Temporal Action Localization with PseudoFormer (CVPR 2025 · similarity 0.92)
- PivoTAL: Prior-Driven Supervision for Weakly-Supervised Temporal Action Localization (CVPR 2023 · similarity 0.92)
- Similar Modality Enhancement and Action Consistency Learning for Weakly Supervised Temporal Action Localization (AAAI 2025 · similarity 0.91)
- Fine-Grained Temporal Contrastive Learning for Weakly-Supervised Temporal Action Localization (CVPR 2022 · similarity 0.91)
- Weakly Supervised Temporal Action Localization via Dual-Prior Collaborative Learning Guided by Multimodal Large Language Models (CVPR 2025 · similarity 0.90)
- Weakly-Supervised Temporal Action Localization by Inferring Salient Snippet-Feature (AAAI 2024 · similarity 0.90)
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.