Research map / Label-efficient and robust learning / Domain adaptation and generalization
Action Segmentation With Joint Self-Supervised Temporal Domain Adaptation
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 local binary domain prediction to MS-TCN | MS-TCN · GTEA | F1@25: 83.6 -> 87.9 | Helped |
| Add joint local and global domain prediction | MS-TCN with Local SSTDA · 50Salads | F1@25: 77.8 -> 81.5 | Helped |
| Use SSTDA rather than source-only training | MS-TCN · Breakfast | F1@25: 59.6 -> 69.1 | Helped |
| Train SSTDA with 65% of labeled data instead of all labeled data | SSTDA · GTEA | F1@25: 89.1 -> 82.6 | Hurt |
| Integrate binary domain prediction at S1 rather than the S2/S3 configuration (ablation) | Local SSTDA · GTEA | F1@25: 87.9 -> 86.2 | Hurt |
| Integrate binary domain prediction at S4 rather than the S2/S3 configuration (ablation) | Local SSTDA · GTEA | F1@25: 87.9 -> 86.4 | Hurt |
| Add a domain predictor at S1 alongside S2 (ablation) | Local SSTDA at S2 · GTEA | F1@25: 87.2 -> 85.8 | Hurt |
| Use 1 segment instead of 2 for sequential domain prediction (ablation) | SSTDA · GTEA | F1@25: 89.1 -> 87.7 | Hurt |
| Use 3 segments instead of 2 for sequential domain prediction (ablation) | SSTDA · GTEA | F1@25: 89.1 -> 87.6 | Hurt |
| Use within-domain temporal shuffling rather than cross-domain temporal prediction | Video self-supervised adaptation · 50Salads and Breakfast | action segmentation performance | Hurt |
Closest papers by meaning
- Unsupervised Action Segmentation via Fast Learning of Semantically Consistent Actoms (AAAI 2024 · similarity 0.89)
- FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Action Segmentation (CVPR 2024 · similarity 0.88)
- OnlineTAS: An Online Baseline for Temporal Action Segmentation (NeurIPS 2024 · similarity 0.87)
- Learning Causal Domain-Invariant Temporal Dynamics for Few-Shot Action Recognition (ICML 2024 · similarity 0.87)
- Unsupervised Video Domain Adaptation with Masked Pre-Training and Collaborative Self-Training (CVPR 2024 · similarity 0.87)
- HopaDIFF: Holistic-Partial Aware Fourier Conditioned Diffusion for Referring Human Action Segmentation in Multi-Person Scenarios (NeurIPS 2025 · 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.