Research map / Label-efficient and robust learning / Domain adaptation and generalization
MC-PanDA: Mask Confidence for Panoptic Domain Adaptation
Ivan Martinović, Josip Šarić, Siniša Šegvić
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 mask-wide loss scaling (MLS) to baseline consistency (ablation) | MC-PanDA · Synthia->Cityscapes | PQ16: 39.6 -> 44.6 | Improved |
| Add confidence-based point filtering (CBPF) on top of MLS (ablation) | MC-PanDA · Synthia->Cityscapes | PQ16: 44.6 -> 47.4 | Improved |
| Add mean-teacher consistency to supervised Synthia baseline (ablation) | Mask2Former · Synthia->Cityscapes | PQ16: 30.8 -> 39.6 | Improved |
| Use mask-wide instead of image-wide loss scaling (ablation) | MC-PanDA · Synthia->Cityscapes | PQ16: 42.1 -> 47.4 | Improved |
| Use all-masks teacher confidence instead of per-mask in CBPF (ablation) | MC-PanDA · Synthia->Cityscapes | PQ16: 44.7 -> 47.4 | Improved |
| Use CBPF instead of random point sampling (ablation) | MC-PanDA · Synthia->Cityscapes | PQ16: 44.2 -> 47.4 | Improved |
| Use tau1=0.99 instead of 0.968 for MLS with CBPF (ablation) | MC-PanDA · Synthia->Cityscapes | PQ16: 47.2 -> 47.4 | No clear effect |
| Use stricter tau2=0.9 for point filtering (ablation) | MC-PanDA · Synthia->Vistas | PQ16: 37.9 -> 38.7 | Improved |
| Set tau1 for stuff classes to 0.9 (ablation) | MC-PanDA · Synthia->Vistas | PQ16: 38.7 -> 39.6 | Improved |
| Use MC-PanDA instead of EDAPS with the same MiT-B5 backbone | MC-PanDA · Synthia->Cityscapes | PQ16: 41.2 -> 44.7 | Improved |
Later papers that compared with this method
1 later accepted paper reports MC-PanDA as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- MC-PanDA++: Simpler, Stronger, and More Robust Domain-Adaptive Panoptic Segmentation (IJCV 2026 · better on 6 of 6 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- EDAPS — EDAPS: Enhanced Domain-Adaptive Panoptic Segmentation (ICCV 2023) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- UniDAformer: Unified Domain Adaptive Panoptic Segmentation Transformer via Hierarchical Mask Calibration (CVPR 2023 · similarity 0.90)
- EDAPS: Enhanced Domain-Adaptive Panoptic Segmentation (ICCV 2023 · similarity 0.89)
- MC-PanDA++: Simpler, Stronger, and More Robust Domain-Adaptive Panoptic Segmentation (IJCV 2026 · similarity 0.89)
- ReMaX: Relaxing for Better Training on Efficient Panoptic Segmentation (NeurIPS 2023 · similarity 0.88)
- Prior2Former - Evidential Modeling of Mask Transformers for Assumption-Free Open-World Panoptic Segmentation (ICCV 2025 · similarity 0.88)
- MaskTwins: Dual-form Complementary Masking for Domain-Adaptive Image Segmentation (ICML 2025 · similarity 0.86)
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.
