Research map / Detection and segmentation / Weakly supervised detection and segmentation
Toward Joint Thing-and-Stuff Mining for Weakly Supervised Panoptic Segmentation
Check 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 |
|---|---|---|---|
| Replace rectangular pooling with shape-invariant MoIPool (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 36.5 -> 39.0 | Helped |
| Use RoIAlign instead of shape-invariant MoIPool (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 39.0 -> 36.2 | Hurt |
| Use shape-interpolation rather than shape-invariant MoIPool (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 39.0 -> 37.2 | Hurt |
| Add instance segmentation refinement only (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 30.7 -> 34.9 | Helped |
| Add semantic segmentation refinement only (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 30.7 -> 33.9 | Helped |
| Add both instance and semantic segmentation refinement (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 30.7 -> 36.5 | Helped |
| Add bottom-up mask re-estimation to LIS and LSS (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 36.5 -> 39.0 | Helped |
| Add one panoptic refinement head rather than none (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 30.3 -> 35.0 | Helped |
| Increase panoptic refinement heads from 1 to 2 (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 35.0 -> 36.5 | Helped |
| Increase panoptic refinement heads from 2 to 3 (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 36.5 -> 38.1 | Helped |
| Increase panoptic refinement heads from 3 to 4 (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 38.1 -> 39.0 | Helped |
| Increase panoptic refinement heads from 4 to 5 (ablation) | JTSM · PASCAL VOC 2012 panoptic segmentation | PQ: 39.0 -> 38.6 | Hurt |
Closest papers by meaning
- Unifying Training and Inference for Panoptic Segmentation (CVPR 2020 · similarity 0.89)
- LPSNet: A Lightweight Solution for Fast Panoptic Segmentation (CVPR 2021 · similarity 0.88)
- Revisiting Open-Set Panoptic Segmentation (AAAI 2024 · similarity 0.87)
- Scene-Centric Unsupervised Panoptic Segmentation (CVPR 2025 · similarity 0.86)
- Fully Data-Driven Pseudo Label Estimation for Pointly-Supervised Panoptic Segmentation (AAAI 2024 · similarity 0.86)
- WISH: Weakly Supervised Instance Segmentation using Heterogeneous Labels (CVPR 2025 · similarity 0.85)
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.