Research map / Detection and segmentation / CNN features
SGN: Sequential Grouping Networks for Instance Segmentation
Shu Liu, Jiaya Jia, Sanja Fidler, Raquel Urtasun
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 MergerNet on top of PSP semantic segmentation (ablation) | SGN · Cityscapes val | AP: 25.2 -> 29.2 | Improved |
| add MergerNet on top of LRR semantic segmentation (ablation) | SGN · Cityscapes val | AP: 20.0 -> 21.7 | Improved |
| use PSP (ResNet-101) instead of LRR (VGG-16) semantic segmentation (ablation) | SGN · Cityscapes val | AP: 21.7 -> 29.2 | Improved |
| use LineNet instead of connected-component baseline (ablation) | SGN · Cityscapes val | AP: 22.1 -> 25.2 | Improved |
| use LineNet instead of heuristic line merging (ablation) | SGN · Cityscapes val | AP: 23.6 -> 25.2 | Improved |
| use heuristic line segment merging instead of plain connected components (ablation) | SGN · Cityscapes val | AP: 22.1 -> 23.6 | Improved |
| increase LineNet history/future context rows (h1f1 to h5f5) (ablation) | LineNet · Cityscapes val | AP: 25.3 -> 25.0 | No clear effect |
| raise MergerNet max distance from 50/100 to 150 pixels (ablation) | MergerNet · Cityscapes val | AP | Hurt |
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.
- InstanceCut — InstanceCut: From Edges to Instances With MultiCut (CVPR 2017) · better on 1 of 1 shared result
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
- SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentation (CVPR 2018 · similarity 0.89)
- Recurrent Pixel Embedding for Instance Grouping (CVPR 2018 · similarity 0.88)
- Pixelwise Instance Segmentation With a Dynamically Instantiated Network (CVPR 2017 · similarity 0.87)
- Instance Segmentation in 3D Scenes Using Semantic Superpoint Tree Networks (ICCV 2021 · similarity 0.87)
- Semantic-aware SAM for Point-Prompted Instance Segmentation (CVPR 2024 · similarity 0.86)
- Training-Free Dataset Pruning for Instance Segmentation (ICLR 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.
