Research map / Point clouds / Point cloud classification
Event-LiDAR: 3D Eventification for Efficient Point Cloud Processing
Masatoshi Murakami, Eisho Tsuji, Ken Sakurada
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 temporal eventification with uniform random sampling at about 80% reduction (ablation) | PTv3 10-scan · nuScenes val | mAP: 63.5 -> 53.2 | Hurt |
| add 3D eventification (10T) instead of random sampling | PTv3 10-scan · nuScenes val | mAP: 53.2 -> 60.9 | Improved |
| keep the current scan full and eventify past scans (1F+9T instead of 10T) | PTv3 on nuScenes · nuScenes val | mAP: 60.9 -> 62.9 | Improved |
| keep the current scan full and eventify past scans (1F+2T instead of 3T) | PTv3 on Waymo · Waymo val | L2 mAP: 67.0 -> 68.4 | Improved |
| add the temporal residual feature delta to the point features (ablation) | PTv3 10T · nuScenes val | mAP: 60.9 -> 61.8 | Improved |
| add the temporal residual feature delta to the point features (ablation) | PTv3 1F+9T · nuScenes val | mAP: 62.9 -> 63.1 | Improved |
| add the temporal residual feature delta (ablation) | CenterPoint 1F+9T · nuScenes val | NDS: 66.1 -> 65.5 | Hurt |
| add the event-aware network design (more early-stage capacity) (ablation) | PTv3 10F · nuScenes val | mAP: 62.3 -> 63.5 | Improved |
| add the event-aware network design (more early-stage capacity) (ablation) | PTv3 1F+9T · nuScenes val | mAP: 62.0 -> 63.1 | Improved |
| use line-based distance instead of point-based distance (ablation) | CenterPoint 10T · nuScenes val | mAP: 55.8 -> 58.7 | Improved |
| use plane-based distance instead of line-based distance (ablation) | CenterPoint 10T · nuScenes val | mAP: 58.7 -> 57.5 | Hurt |
| use plane-based distance instead of line-based distance (ablation) | CenterPoint 1F+2T · Waymo val | L2 mAP: 67.1 -> 67.4 | Improved |
Closest papers by meaning
- MGTANet: Encoding Sequential LiDAR Points Using Long Short-Term Motion-Guided Temporal Attention for 3D Object Detection (AAAI 2023 · similarity 0.89)
- iDet3D: Towards Efficient Interactive Object Detection for LiDAR Point Clouds (AAAI 2024 · similarity 0.88)
- PTT: Point-Trajectory Transformer for Efficient Temporal 3D Object Detection (CVPR 2024 · similarity 0.87)
- TREND: Unsupervised 3D Representation Learning via Temporal Forecasting for LiDAR Perception (NeurIPS 2025 · similarity 0.87)
- LiDAR-Net: A Real-scanned 3D Point Cloud Dataset for Indoor Scenes (CVPR 2024 · similarity 0.87)
- An Empirical Study of the Generalization Ability of Lidar 3D Object Detectors to Unseen Domains (CVPR 2024 · similarity 0.86)
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.
