Research map / Point clouds / Segmentation and detection
LoGoNet: Towards Accurate 3D Object Detection With Local-to-Global Cross-Modal Fusion
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 |
|---|---|---|---|
| Use local-to-global fusion rather than the LiDAR-only baseline | LoGoNet on Waymo · Waymo val | mAPH (L2): +0.81%, +2.93%, and +2.25% on vehicle, pedestrian, and cyclist | Helped |
| Use local-to-global fusion rather than the LiDAR-only baseline | LoGoNet on KITTI · KITTI val | mAP: +0.70%, +4.83%, and +3.66% on car, pedestrian, and cyclist | Helped |
| Add GoF with centroid-based global fusion to single-modal refinement (ablation) | LoGoNet on Waymo · Waymo val | APH L2: 0.97%, 1.68% and 0.97% | Helped |
| Add LoF to single-modal refinement (ablation) | LoGoNet on Waymo · Waymo val | APH L2: +1.22%, +2.93%, and +1.50% | Helped |
| Add FDA to the GoF+LoF model (ablation) | LoGoNet on Waymo · Waymo val | APH L2: 0.55%, 0.25% and 0.46% | Helped |
| Combine GoF and LoF without FDA (ablation) | LoGoNet on Waymo · Waymo val, cyclist | APH L2: 69.23 -> 68.90 | Hurt |
| Add point-centroid information to XYZ+D grid position features (ablation) | LoF on Waymo · Waymo val | APH L2: 0.12%, 0.28%, and 0.15% | Helped |
| Use LoGoNet with 5-frame input instead of MPPNet with 16-frame input | Waymo detection · Waymo val | mAPH: 74.85 -> 75.54 | Helped |
Closest papers by meaning
- ViKIENet: Towards Efficient 3D Object Detection with Virtual Key Instance Enhanced Network (CVPR 2025 · similarity 0.89)
- SimDistill: Simulated Multi-Modal Distillation for BEV 3D Object Detection (AAAI 2024 · similarity 0.89)
- PromptDet: A Lightweight 3D Object Detection Framework with LiDAR Prompts (AAAI 2025 · similarity 0.88)
- SAFDNet: A Simple and Effective Network for Fully Sparse 3D Object Detection (CVPR 2024 · similarity 0.88)
- Voxel or Pillar: Exploring Efficient Point Cloud Representation for 3D Object Detection (AAAI 2024 · similarity 0.88)
- GBlobs: Explicit Local Structure via Gaussian Blobs for Improved Cross-Domain LiDAR-based 3D Object Detection (CVPR 2025 · similarity 0.87)
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.