Research map / Detection and segmentation / CNN features
GP-S3Net: Graph-Based Panoptic Sparse Semantic Segmentation Network
Ryan Razani, Ran Cheng, Enxu Li, Ehsan Taghavi, Yuan Ren, Liu Bingbing
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 |
|---|---|---|---|
| replace DBSCAN clustering with the GNN edge-classification instance network (ablation) | semantic backbone with clustering · SemanticKITTI val | PQ: 49.4 -> 63.3 | Improved |
| replace HDBSCAN clustering with the GNN edge-classification instance network (ablation) | semantic backbone with clustering · SemanticKITTI val | PQ: 53.4 -> 63.3 | Improved |
| replace MeanShift clustering with the GNN edge-classification instance network (ablation) | semantic backbone with clustering · SemanticKITTI val | PQ: 55.8 -> 63.3 | Improved |
| replace MeanShift clustering with the GNN edge-classification instance network (ablation) | semantic backbone with clustering · SemanticKITTI val | PQ Th: 55.7 -> 70.2 | Improved |
| use MeanShift instead of HDBSCAN as clustering baseline (ablation) | semantic backbone with clustering · SemanticKITTI val | PQ: 53.4 -> 55.8 | Improved |
| use HDBSCAN instead of DBSCAN as clustering baseline (ablation) | semantic backbone with clustering · SemanticKITTI val | PQ: 49.4 -> 53.4 | Improved |
| use GP-S3Net instead of DS-Net | DS-Net · SemanticKITTI test | PQ: 55.9 -> 60.0 | Improved |
| use GP-S3Net instead of Panoster | Panoster · SemanticKITTI test | PQ: 52.7 -> 60.0 | Improved |
| use GP-S3Net instead of EfficientLPS | EfficientLPS · SemanticKITTI test | PQ: 57.4 -> 60.0 | Improved |
| use GP-S3Net instead of EfficientLPS | EfficientLPS · SemanticKITTI test | PQ St: 60.5 -> 56.4 | Hurt |
| use GP-S3Net instead of EfficientLPS | EfficientLPS · nuScenes val | PQ: 59.2 -> 61.0 | Improved |
| use GP-S3Net instead of DS-Net | DS-Net · SemanticPOSS val | PQ: 35.6 -> 48.7 | Improved |
Later papers that compared with this method
1 later accepted paper reports GP-S3Net as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Sparse Cross-Scale Attention Network for Efficient LiDAR Panoptic Segmentation (AAAI 2022 · 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
- Panoptic-PolarNet: Proposal-Free LiDAR Point Cloud Panoptic Segmentation (CVPR 2021 · similarity 0.92)
- LiDAR-Based Panoptic Segmentation via Dynamic Shifting Network (CVPR 2021 · similarity 0.91)
- Panoptic-PHNet: Towards Real-Time and High-Precision LiDAR Panoptic Segmentation via Clustering Pseudo Heatmap (CVPR 2022 · similarity 0.91)
- Bidirectional Graph Reasoning Network for Panoptic Segmentation (CVPR 2020 · similarity 0.90)
- Sparse Cross-Scale Attention Network for Efficient LiDAR Panoptic Segmentation (AAAI 2022 · similarity 0.90)
- (AF)2-S3Net: Attentive Feature Fusion With Adaptive Feature Selection for Sparse Semantic Segmentation Network (CVPR 2021 · similarity 0.90)
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.
