Research map / Detection and segmentation / Weakly supervised detection and segmentation
How to Save your Annotation Cost for Panoptic 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 Pareto-optimal ratio O6 rather than random ratio R6 (ablation) | CQB-Net · COCO val | PQall: 41.7 -> 42.3 | Helped |
| Use Pareto-optimal ratio O6 rather than random ratio R6 (ablation) | CQB-Net · COCO val | annotation cost: 0.570 -> 0.423 | Hurt |
| Use Pareto-optimal ratio O7 with less than half the full annotation cost | CQB-Net · COCO val | PQall: 42.5 -> 42.7 | Helped |
| Use Pareto-optimal ratio O6 with ResNet-101 | CQB-Net · COCO val | PQall: 43.0 -> 43.4 | Helped |
| Add the Cls2Box module (ablation) | CQB-Net · COCO val, ρP=0.1 | PQth: 40.1 -> 44.0 | Helped |
| Add the Box2Mask module (ablation) | CQB-Net · COCO val, ρP=0.1 | PQth: 47.3 -> 48.8 | Helped |
| Add stuff-to-thing relation reasoning (ablation) | CQB-Net · COCO val, ρP=0.1 | PQth: 39.1 -> 39.7 | Helped |
| Add the Cls2Pixel module (ablation) | CQB-Net · COCO val, ρP=0.1 | PQst: 25.3 -> 26.5 | Helped |
| Add thing-to-stuff relation reasoning (ablation) | CQB-Net · COCO val, ρP=0.1 | PQst: 24.4 -> 26.1 | Helped |
Closest papers by meaning
- Ada-Segment: Automated Multi-loss Adaptation for Panoptic Segmentation (AAAI 2021 · similarity 0.87)
- Fully Data-Driven Pseudo Label Estimation for Pointly-Supervised Panoptic Segmentation (AAAI 2024 · similarity 0.86)
- COCONut-PanCap: Joint Panoptic Segmentation and Grounded Captions for Fine-Grained Understanding and Generation (NeurIPS 2025 · similarity 0.84)
- Versatile Medical Image Segmentation Learned from Multi-Source Datasets via Model Self-Disambiguation (CVPR 2024 · similarity 0.84)
- Revisiting Open-Set Panoptic Segmentation (AAAI 2024 · similarity 0.84)
- Stratify or Die: Rethinking Data Splits in Image Segmentation (NeurIPS 2025 · similarity 0.84)
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.