Research map / Label-efficient and robust learning / Domain adaptation and generalization
When Visual Prompt Tuning Meets Source-Free Domain Adaptive Semantic Segmentation
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 |
|---|---|---|---|
| add level embedding to prompt generator (ablation) | prompt adapter · GTA5 -> Cityscapes | mIoU (%): 55.58 -> 56.24 | Helped |
| use multiscale feature maps instead of single-scale (ablation) | prompt generator · GTA5 -> Cityscapes | mIoU (%): 55.52 -> 56.24 | Helped |
| add stem module (ablation) | prompt generator · GTA5 -> Cityscapes | mIoU (%): 55.50 -> 56.24 | Helped |
| use multiple prompt interactors instead of only PI 1 (ablation) | prompt adapter · GTA5 -> Cityscapes | mIoU (%): 55.34 -> 56.24 | Helped |
| add the prompt adapter (generator and interactors) (ablation) | Uni-UVPT · GTA5 -> Cityscapes | mIoU (%): 55.07 -> 56.24 | Helped |
| adaptive pseudo-label correction instead of ELR (ablation) | Uni-UVPT · GTA5 -> Cityscapes | mIoU (%): 55.60 -> 56.24 | Helped |
| correct pseudo labels adaptively rather than once offline (ablation) | Uni-UVPT · GTA5 -> Cityscapes | mIoU (%): 55.47 -> 56.24 | Helped |
| add feature consistency loss (ablation) | multiscale consistency loss · GTA5 -> Cityscapes | mIoU (%): 54.26 -> 56.24 | Helped |
| add prediction consistency loss (ablation) | multiscale consistency loss · GTA5 -> Cityscapes | mIoU (%): 56.01 -> 56.24 | Helped |
| add both consistency terms (ablation) | Uni-UVPT · GTA5 -> Cityscapes | mIoU (%): 53.81 -> 56.24 | Helped |
| weather augmentation instead of scale augmentation (Swin-B, standard) | Uni-UVPT · GTA5 -> Cityscapes | mIoU (%): 56.2 -> 54.7 | Hurt |
| weather augmentation instead of scale augmentation (Swin-B, standard) | Uni-UVPT · SYNTHIA -> Cityscapes (16 classes) | mIoU (%): 52.6 -> 52.9 | Helped |
Closest papers by meaning
- Domain-Agnostic Mutual Prompting for Unsupervised Domain Adaptation (CVPR 2024 · similarity 0.90)
- SA²VP: Spatially Aligned-and-Adapted Visual Prompt (AAAI 2024 · similarity 0.89)
- Exploring Sparse Visual Prompt for Domain Adaptive Dense Prediction (AAAI 2024 · similarity 0.89)
- Prompt-Based Distribution Alignment for Unsupervised Domain Adaptation (AAAI 2024 · similarity 0.88)
- DA-VPT: Semantic-Guided Visual Prompt Tuning for Vision Transformers (CVPR 2025 · similarity 0.88)
- Preserving Clusters in Prompt Learning for Unsupervised Domain Adaptation (CVPR 2025 · similarity 0.88)
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.