Research map / Detection and segmentation / 3D detection and driving
LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving
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 |
|---|---|---|---|
| Predict a probability distribution over boxes rather than only the mean (ablation) | LaserNet · ATG4D vehicle detection | Vehicle AP0.7: 77.05 -> 85.34 | Helped |
| Map range-image rows using LiDAR laser IDs rather than uniform elevation spacing (ablation) | LaserNet · ATG4D vehicle detection | Vehicle AP0.7: 79.14 -> 80.92 | Helped |
| Add mean shift clustering to per-point box predictions (ablation) | LaserNet with unimodal predictions and fixed Hard NMS · ATG4D vehicle detection | Vehicle AP0.7: 80.22 -> 80.92 | Helped |
| Use a multimodal rather than unimodal box distribution (ablation) | LaserNet with Laser spacing, mean shift, and fixed Hard NMS · ATG4D vehicle detection | Vehicle AP0.7: 80.92 -> 81.80 | Helped |
| Use adaptive Soft NMS rather than adaptive Hard NMS (ablation) | LaserNet · ATG4D vehicle detection | Vehicle AP0.7: 83.68 -> 85.34 | Helped |
| Use adaptive Soft NMS rather than Soft NMS with no fixed IoU threshold (ablation) | LaserNet · ATG4D vehicle detection | Vehicle AP0.7: 84.43 -> 85.34 | Helped |
| Operate in native range view with a compact dense input | LaserNet · KITTI runtime | Forward Pass (ms): 35 -> 12 | Helped |
| Operate in native range view with a compact dense input | LaserNet · KITTI runtime | Total (ms): 62 -> 30 | Helped |
| Train the probabilistic range-view detector on a small rather than large dataset | LaserNet · KITTI vehicle detection | Vehicle AP0.7 | Hurt |
Closest papers by meaning
- Pseudo-LiDAR From Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving (CVPR 2019 · similarity 0.92)
- Multi-View 3D Object Detection Network for Autonomous Driving (CVPR 2017 · similarity 0.92)
- RangeIoUDet: Range Image Based Real-Time 3D Object Detector Optimized by Intersection Over Union (CVPR 2021 · similarity 0.91)
- Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data (CVPR 2022 · similarity 0.91)
- SAFDNet: A Simple and Effective Network for Fully Sparse 3D Object Detection (CVPR 2024 · similarity 0.89)
- Pre-training LiDAR-based 3D Object Detectors through Colorization (ICLR 2024 · similarity 0.89)
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.