Research map / Point clouds / Segmentation and detection
Multimodal Virtual Point 3D Detection
Read the paperCheck 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 multimodal virtual points (ablation) | CenterPoint VoxelNet · nuScenes val | mAP: 59.6 -> 65.9 | Helped |
| add multimodal virtual points (ablation) | CenterPoint PointPillars · nuScenes val | mAP: 52.3 -> 62.7 | Helped |
| add split voxelization (ablation) | MVP VoxelNet · nuScenes val | NDS: 69.6 -> 70.0 | Helped |
| add two-stage refinement (ablation) | MVP VoxelNet · nuScenes val | mAP: 66.0 -> 67.1 | Helped |
| add two-stage refinement (ablation) | CenterPoint VoxelNet without virtual points · nuScenes val | mAP: 59.6 -> 60.5 | Helped |
| add virtual points on top of point painting (ablation) | CenterPoint + painting · nuScenes val 30-50m | mAP: 46.2 -> 47.3 | Helped |
| add virtual points on top of point painting (ablation) | CenterPoint + painting · nuScenes val 0-15m | mAP: 78.2 -> 78.1 | No clear effect |
| reduce 2D segmentation input resolution 900 -> 480 (ablation) | MVP · nuScenes val | NDS: 70.0 -> 69.2 | Hurt |
| add multimodal virtual points | PointPillars · KITTI moderate | 3D mAP Cyclist: 62.7 -> 65.0 | Helped |
| add multimodal virtual points | PointPillars · KITTI moderate | 3D mAP Pedestrian: 52.3 -> 50.5 | Hurt |
| add multimodal virtual points | CenterPoint · nuScenes test | mAP: 58.0 -> 66.4 | Helped |
Closest papers by meaning
- Multi-View 3D Object Detection Network for Autonomous Driving (CVPR 2017 · similarity 0.91)
- LiDAR R-CNN: An Efficient and Universal 3D Object Detector (CVPR 2021 · similarity 0.90)
- ViKIENet: Towards Efficient 3D Object Detection with Virtual Key Instance Enhanced Network (CVPR 2025 · similarity 0.90)
- Pre-training LiDAR-based 3D Object Detectors through Colorization (ICLR 2024 · similarity 0.90)
- Beyond Virtual Points: Depth-Enhanced LiDAR-only 3D Object Detection with Semi-Supervised Learning (Student Abstract) (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.