Research map / Video / Action recognition
Learning a Weakly-Supervised Video Actor-Action Segmentation Model With a Wise Selection
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 |
|---|---|---|---|
| Train on selected high-quality PAs rather than the full PA set (ablation) | WS² · A2D test | mIoU actor-action: 20.38 -> 24.84 | Helped |
| Select the model for PA updates using RIC rather than mIoU_PA (ablation) | WS² with selected PAs · A2D test | mIoU actor-action: 24.84 -> 26.74 | Helped |
| Use RIC rather than mIoU_PA to select the next PA version (ablation) | WS² · A2D PA.v1 actor-action PAs | mIoU_GT: 28.58 -> 29.27 | Helped |
| Use RIC rather than mIoU_PA to select the next PA version (ablation) | WS² · A2D PA.v2 actor-action PAs | mIoU_GT: 31.72 -> 32.36 | Helped |
| Use RIC rather than mIoU_PA to select the next PA version (ablation) | WS² · A2D PA.v3 actor-action PAs | mIoU_GT: 33.05 -> 33.64 | Helped |
| Use selected PAs and RIC-based model selection together instead of full PAs and mIoU_PA selection (ablation) | WS² · A2D test | mIoU actor-action: 20.38 -> 26.74 | Helped |
| Use WS² rather than the strongest listed prior video-level weakly-supervised method | YouTube-Object · YouTube-Object test | mIoU: 63.1 -> 64.7 | Helped |
Closest papers by meaning
- Weakly Supervised Actor-Action Segmentation via Robust Multi-Task Ranking (CVPR 2017 · similarity 0.91)
- Action Sets: Weakly Supervised Action Segmentation Without Ordering Constraints (CVPR 2018 · similarity 0.90)
- Efficient and Effective Weakly-Supervised Action Segmentation via Action-Transition-Aware Boundary Alignment (CVPR 2024 · similarity 0.88)
- SCT: Set Constrained Temporal Transformer for Set Supervised Action Segmentation (CVPR 2020 · similarity 0.88)
- ASM-Loc: Action-Aware Segment Modeling for Weakly-Supervised Temporal Action Localization (CVPR 2022 · similarity 0.88)
- Semi-Weakly-Supervised Learning of Complex Actions From Instructional Task Videos (CVPR 2022 · 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.