Research map / Video / Action recognition
Modeling Multi-Label Action Dependencies for Temporal Action Localization
Check 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 |
|---|---|---|---|
| Model co-occurrence and temporal action dependencies with MLAD attention | I3D-based temporal action localization · MultiTHUMOS | f-mAP: 48.2 -> 51.5 | Helped |
| Model co-occurrence and temporal action dependencies with MLAD attention | I3D-based temporal action localization · Charades | f-mAP: 22.9 -> 23.7 | Helped |
| Add 5 MLAD layers to class-level features (ablation) | CF Baseline · MultiTHUMOS | f-mAP: 42.6 -> 51.52 | Helped |
| Add 5 MLAD layers to class-level features (ablation) | CF Baseline · Charades | f-mAP: 14.8 -> 23.74 | Helped |
| Increase MLAD depth from 1 to 3 layers (ablation) | MLAD action localizer · MultiTHUMOS | f-mAP: 48.55 -> 50.30 | Helped |
| Increase MLAD depth from 3 to 5 layers (ablation) | MLAD action localizer · MultiTHUMOS | f-mAP: 50.30 -> 51.52 | Helped |
| Increase MLAD depth from 5 to 7 layers (ablation) | MLAD action localizer · MultiTHUMOS | f-mAP | No clear effect |
| Vary sequence length during training rather than fixing it (ablation) | MLAD action localizer · MultiTHUMOS, evaluation T=128 | f-mAP: 50.59 -> 51.52 | Helped |
| Add the Co-occurrence Dependency Branch (ablation) | Class-level feature baseline · MultiTHUMOS | f-mAP: 42.60 -> 44.98 | Helped |
| Add the Temporal Dependency Branch (ablation) | Class-level feature baseline · MultiTHUMOS | f-mAP: 42.60 -> 48.03 | Helped |
| Use both Co-occurrence and Temporal Dependency Branches (ablation) | MLAD action localizer · MultiTHUMOS | f-mAP: 48.03 -> 51.52 | Helped |
| Use both Co-occurrence and Temporal Dependency Branches (ablation) | MLAD action localizer · Charades | f-mAP: 21.1 -> 23.5 | Helped |
Closest papers by meaning
- Learning Temporal Co-Attention Models for Unsupervised Video Action Localization (CVPR 2020 · similarity 0.91)
- Learning To Refactor Action and Co-Occurrence Features for Temporal Action Localization (CVPR 2022 · similarity 0.91)
- Fine-Grained Temporal Contrastive Learning for Weakly-Supervised Temporal Action Localization (CVPR 2022 · similarity 0.91)
- PointTAD: Multi-Label Temporal Action Detection with Learnable Query Points (NeurIPS 2022 · 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 0.90)
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.