Research map / Detection and segmentation / 3D detection and driving
A Versatile Multi-View Framework for LiDAR-Based 3D Object Detection With Guidance From 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 |
|---|---|---|---|
| Add panoptic segmentation guidance to CenterPoint | CenterPoint · nuScenes validation | mAP: 56.4 -> 60.3 | Helped |
| Add panoptic segmentation guidance to CenterPoint | CenterPoint · nuScenes validation | NDS: 64.8 -> 67.1 | Helped |
| Add panoptic segmentation guidance to CenterPoint | CenterPoint · nuScenes test | mAP: 58.0 -> 60.9 | Helped |
| Add panoptic segmentation guidance to CenterPoint | CenterPoint · nuScenes test | NDS: 65.5 -> 67.3 | Helped |
| Add multi-view backbone augmentation (MBA) (ablation) | CenterPoint · nuScenes validation | NDS: 63.8 -> 66.5 | Helped |
| Add class-wise foreground attention (CFA) to MBA (ablation) | CenterPoint with MBA · nuScenes validation | NDS: 66.5 -> 66.9 | Helped |
| Add center density heatmap (CDH) to MBA + CFA (ablation) | CenterPoint with MBA + CFA · nuScenes validation | NDS: 66.9 -> 67.1 | Helped |
| Apply the multi-task framework | PointPillars · nuScenes validation | mAP: 43.0 -> 50.5 | Helped |
| Apply the multi-task framework | PointPillars · nuScenes validation | NDS: 56.8 -> 60.5 | Helped |
| Apply the multi-task framework | SECOND · nuScenes validation | mAP: 51.7 -> 56.2 | Helped |
| Apply the multi-task framework | SECOND · nuScenes validation | NDS: 62.6 -> 64.8 | Helped |
| Jointly train panoptic segmentation and detection rather than freeze pre-trained CPSeg (ablation) | CenterPoint + CPSeg · nuScenes validation | mAP: 59.9 -> 60.3 | Helped |
Closest papers by meaning
- Multi-View 3D Object Detection Network for Autonomous Driving (CVPR 2017 · similarity 0.91)
- Fully Convolutional One-Stage 3D Object Detection on LiDAR Range Images (NeurIPS 2022 · similarity 0.90)
- VISTA: Boosting 3D Object Detection via Dual Cross-VIew SpaTial Attention (CVPR 2022 · similarity 0.90)
- GeoBEV: Learning Geometric BEV Representation for Multi-view 3D Object Detection (AAAI 2025 · similarity 0.89)
- Towards Generalizable Multi-Camera 3D Object Detection via Perspective Rendering (AAAI 2025 · similarity 0.89)
- SimDistill: Simulated Multi-Modal Distillation for BEV 3D Object Detection (AAAI 2024 · 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.