Research map / Video / Action recognition
Set-Supervised Action Learning in Procedural Task Videos via Pairwise Order Consistency
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 |
|---|---|---|---|
| Add pairwise ordering consistency loss (ablation) | POC action segmentation model · Breakfast | MoF: 37.0 -> 40.1 | Helped |
| Add pairwise ordering consistency loss (ablation) | POC action segmentation model · Breakfast | IoU: 30.3 -> 32.5 | Helped |
| Add frame-level ranking loss (ablation) | POC action segmentation model · Breakfast | MoF: 15.5 -> 36.6 | Helped |
| Add frame-level ranking loss (ablation) | POC action segmentation model · Breakfast | IoU: 9.3 -> 28.7 | Helped |
| Add frame-level centering loss (ablation) | POC action segmentation model · Breakfast | MoF: 38.5 -> 40.1 | Helped |
| Add frame-level centering loss (ablation) | POC action segmentation model · Breakfast | IoU: 31.7 -> 32.5 | Helped |
| Use video-level cross-entropy rather than ranking loss alone (ablation) | POC action segmentation model · Breakfast | MoF: 32.5 -> 40.1 | Helped |
| Use ground-truth transcripts to provide reference ordering | POC · Breakfast, action segmentation | IoU: 33.5 -> 39.4 | Helped |
| Use ground-truth transcripts to provide reference ordering | POC · CrossTask, action segmentation | MoF: 44.0 -> 44.1 | No clear effect |
| Use POC rather than SCT | Set-supervised action segmentation · Breakfast | MoF: 34.8 -> 42.4 | Helped |
| Use POC rather than SCT | Set-supervised action segmentation · CrossTask | IoU: 11.4 -> 16.2 | Helped |
| Use POC rather than TASL | Transcript-supervised action segmentation · CrossTask | IoU: 14.9 -> 16.3 | Helped |
Closest papers by meaning
- Efficient and Effective Weakly-Supervised Action Segmentation via Action-Transition-Aware Boundary Alignment (CVPR 2024 · similarity 0.89)
- Similar Modality Enhancement and Action Consistency Learning for Weakly Supervised Temporal Action Localization (AAAI 2025 · similarity 0.88)
- FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Action Segmentation (CVPR 2024 · similarity 0.87)
- Weakly-Supervised Temporal Action Localization by Inferring Salient Snippet-Feature (AAAI 2024 · similarity 0.87)
- Masked Temporal Interpolation Diffusion for Procedure Planning in Instructional Videos (ICLR 2025 · similarity 0.87)
- Unsupervised Action Segmentation via Fast Learning of Semantically Consistent Actoms (AAAI 2024 · similarity 0.87)
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.