Research map / Detection and segmentation / CNN features
Seamless Scene 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 |
|---|---|---|---|
| use one jointly trained backbone and two task-specific heads instead of separately trained models | Ours Combined · Cityscapes val | PQ: 59.8 -> 60.3 | Helped |
| use one jointly trained backbone and two task-specific heads instead of separately trained models | Ours Combined · Mapillary Vistas val | PQ: 37.2 -> 37.7 | Helped |
| use one jointly trained backbone and two task-specific heads instead of separately trained models | Ours Combined · IDD level 3 | PQ: 47.2 -> 46.9 | Hurt |
| use one jointly trained backbone and two task-specific heads instead of separately trained models | Ours Combined · IDD level 3 | PQ†: 48.8 -> 48.6 | Hurt |
| replace global pooling with fixed-kernel average pooling and replicated-boundary padding (ablation) | DeepLabV3 segmentation head · Cityscapes full-resolution segmentation | mIoU: 72.7% -> 74.5% (+1.8% absolute) | Helped |
| replace global pooling with fixed-kernel average pooling and replicated-boundary padding (ablation) | Cityscapes segmentation model trained with crops · Cityscapes segmentation with 768×768 crops | mIoU: 73.3% -> 74.2% (+0.9% absolute) | Helped |
Closest papers by meaning
- Unifying Training and Inference for Panoptic Segmentation (CVPR 2020 · similarity 0.91)
- Real-Time Panoptic Segmentation From Dense Detections (CVPR 2020 · similarity 0.90)
- LPSNet: A Lightweight Solution for Fast Panoptic Segmentation (CVPR 2021 · similarity 0.90)
- Panoptic, Instance and Semantic Relations: A Relational Context Encoder To Enhance Panoptic Segmentation (CVPR 2022 · similarity 0.89)
- Panoptic Feature Pyramid Networks (CVPR 2019 · similarity 0.89)
- Scene-Centric Unsupervised Panoptic Segmentation (CVPR 2025 · similarity 0.88)
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.