Research map / Video / Action recognition
Weakly Supervised Temporal Action Localization Through Learning Explicit Subspaces for Action and Context
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 |
|---|---|---|---|
| Learn explicit action and context subspaces and use action-subspace scores for localization | WS-TAL · THUMOS14 test | average mAP@IoU 0.3:0.1:0.7: 21.6 -> 30.6 | Helped |
| Use action-subspace scores for TAP evaluation (C3) (ablation) | Base module with UntrimmedNet features · THUMOS14 test | average mAP@IoU 0.3:0.1:0.7: 21.6 -> 22.6 | Helped |
| Use action-subspace scores for TAP generation (C2) (ablation) | Base module with UntrimmedNet features · THUMOS14 test | average mAP@IoU 0.3:0.1:0.7: 21.6 -> 22.8 | Helped |
| Add T-ResM with its auxiliary unsupervised training task (C4) (ablation) | C2+C3 WS-TAL with UntrimmedNet features · THUMOS14 test | average mAP@IoU 0.3:0.1:0.7: 23.9 -> 25.8 | Helped |
| Add T-ResM with its auxiliary unsupervised training task (C4) (ablation) | C2+C3 WS-TAL with I3D features · THUMOS14 test | average mAP@IoU 0.3:0.1:0.7: 28.8 -> 29.8 | Helped |
| Replace Base module proposal generation and evaluation with action-subspace scores (ablation) | WS-TAL with UntrimmedNet features · THUMOS14 test | average mAP@IoU 0.3:0.1:0.7: 21.6 -> 26.3 | Helped |
| Replace Base module proposal generation and evaluation with action-subspace scores (ablation) | WS-TAL with I3D features · THUMOS14 test | average mAP@IoU 0.3:0.1:0.7: 22.4 -> 30.6 | Helped |
| Use the full action-context subspace method | WS-TAL with I3D features · ActivityNet v1.2 validation | average mAP: 23.3 -> 25.5 | Helped |
| Use the full action-context subspace method | WS-TAL with I3D features · ActivityNet v1.3 validation | average mAP: 22.2 -> 23.2 | Helped |
Closest papers by meaning
- ACSNet: Action-Context Separation Network for Weakly Supervised Temporal Action Localization (AAAI 2021 · similarity 0.94)
- Weakly-Supervised Temporal Action Localization by Inferring Salient Snippet-Feature (AAAI 2024 · similarity 0.92)
- Bridge the Gap: From Weak to Full Supervision for Temporal Action Localization with PseudoFormer (CVPR 2025 · similarity 0.90)
- Temporal Action Localization with Cross Layer Task Decoupling and Refinement (AAAI 2025 · similarity 0.90)
- Realigning Confidence with Temporal Saliency Information for Point-Level Weakly-Supervised Temporal Action Localization (CVPR 2024 · similarity 0.90)
- Similar Modality Enhancement and Action Consistency Learning for Weakly Supervised Temporal Action Localization (AAAI 2025 · 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.