Research map / Detection and segmentation / Open-set and few-shot detection
Prior2Former - Evidential Modeling of Mask Transformers for Assumption-Free Open-World Panoptic Segmentation
Sebastian Schmidt, Julius Koerner, Dominik Fuchsgruber, Stefano Gasperini, Federico Tombari, Stephan Günnemann
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 no-object mask filtering (ablation) | P2F · SMIYC Road Anomaly | AP: 40.5 -> 58.6 | Improved |
| add no-object mask filtering (ablation) | P2F · FS L&F Obstacle | AP: 66.4 -> 66.6 | No clear effect |
| add symmetric Dice loss (ablation) | P2F · SMIYC Road Anomaly | FPR95: 72.7 -> 46.3 | Improved |
| add symmetric Dice loss (ablation) | P2F · FS L&F Obstacle | FPR95: 95.9 -> 16.8 | Improved |
| add evidential sampling (ablation) | P2F · SMIYC Road Anomaly | AP: 47.2 -> 58.6 | Improved |
| add evidential sampling (ablation) | P2F · FS L&F Obstacle | AP: 23.1 -> 66.6 | Improved |
| use P2F uncertainty instead of classical prediction uncertainty (ablation) | P2F · FS L&F Obstacle | FPR95: 88.4 -> 16.8 | Improved |
| use P2F uncertainty instead of classical prediction uncertainty (ablation) | P2F · SMIYC Road Anomaly | AP: 48.6 -> 58.6 | Improved |
| replace U3HS with P2F | U3HS · OoDIS benchmark, Mean | AP: 0.19 -> 3.21 | Improved |
| replace U3HS with P2F | U3HS · COCO open-world panoptic (unseen classes) | PQ: 9.62 -> 10.71 | Improved |
| replace U3HS with P2F | U3HS · Cityscapes panoptic (open-world) | PQ: 41.21 -> 45.25 | Improved |
| replace U3HS with P2F | U3HS · Lost & Found unseen (reduced size) | PQ: 7.94 -> 11.22 | Improved |
Closest papers by meaning
- Mask2Anomaly: Mask Transformer for Universal Open-Set Segmentation (TPAMI 2024 · similarity 0.92)
- MC-PanDA: Mask Confidence for Panoptic Domain Adaptation (ECCV 2024 · similarity 0.88)
- Panoptic-PartFormer: Learning a Unified Model for Panoptic Part Segmentation (ECCV 2022 · similarity 0.88)
- Masked-Attention Mask Transformer for Universal Image Segmentation (CVPR 2022 · similarity 0.88)
- Segment Anything, Even Occluded (CVPR 2025 · similarity 0.87)
- SGFormer: Semantic-Geometry Fusion Transformer for Multi-modal 3D Panoptic Segmentation (AAAI 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.
