Research map / Detection and segmentation / Open-set and few-shot detection
Pixel-Level Cycle Association: A New Perspective for 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 |
|---|---|---|---|
| Use PLCA instead of CRST | DeepLab-V2 with ResNet-101 · GTAV → Cityscapes | mIoU: 47.1 -> 47.7 | Helped |
| Use PLCA instead of CRST | DeepLab-V2 with ResNet-101 · SYNTHIA → Cityscapes (16 classes) | mIoU: 43.8 -> 46.8 | Helped |
| Use PLCA instead of CRST | DeepLab-V2 with ResNet-101 · SYNTHIA → Cityscapes (13 classes) | mIoU: 50.1 -> 54.0 | Helped |
| Add Lovász-softmax loss (ablation) | Source-only segmentation model · GTAV → Cityscapes | mIoU: 31.5 -> 34.3 | Helped |
| Add feature association loss (ablation) | Source + Lovász-softmax model · GTAV → Cityscapes | mIoU: 34.3 -> 39.7 | Helped |
| Add prediction association loss (ablation) | Model with feature association · GTAV → Cityscapes | mIoU: 39.7 -> 47.4 | Helped |
| Add label-smoothing regularization (ablation) | Model with feature and prediction association · GTAV → Cityscapes | mIoU: 47.4 -> 47.7 | Helped |
| Replace contrastive association with direct similarity encouragement (ablation) | PLCA · GTAV → Cityscapes | mIoU: 47.7 -> 41.8 | Hurt |
| Remove spatial aggregation (ablation) | PLCA · GTAV → Cityscapes | mIoU: 47.7 -> 45.6 | Hurt |
| Remove spatial aggregation (ablation) | PLCA · SYNTHIA → Cityscapes (16 classes) | mIoU: 46.8 -> 44.6 | Hurt |
Closest papers by meaning
- Multi-source Domain Adaptation for Semantic Segmentation (NeurIPS 2019 · similarity 0.91)
- DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation (CVPR 2022 · similarity 0.90)
- Taking a Closer Look at Domain Shift: Category-Level Adversaries for Semantics Consistent Domain Adaptation (CVPR 2019 · similarity 0.90)
- Cross-Domain Grouping and Alignment for Domain Adaptive Semantic Segmentation (AAAI 2021 · similarity 0.90)
- Pay Attention to Target: Relation-Aware Temporal Consistency for Domain Adaptive Video Semantic Segmentation (AAAI 2024 · similarity 0.89)
- Open-Set Domain Adaptation for Semantic Segmentation (CVPR 2024 · similarity 0.89)
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.