Research map / 3D vision / Depth and stereo
L3-Net: Towards Learning Based LiDAR Localization for Autonomous Driving
Weixin Lu, Yao Zhou, Guowei Wan, Shenhua Hou, Shiyu Song
Check 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 |
|---|---|---|---|
| add RNN temporal smoothing (ablation) | L3-Net · BaylandsToSeafood | horizontal RMS: 0.054 -> 0.050 | Improved |
| add RNN temporal smoothing (ablation) | L3-Net · ColumbiaPark | horizontal RMS: 0.047 -> 0.043 | Improved |
| add RNN temporal smoothing (ablation) | L3-Net · Highway237 | horizontal RMS: 0.053 -> 0.045 | Improved |
| add RNN temporal smoothing (ablation) | L3-Net · MathildaAVE | horizontal RMS: 0.054 -> 0.051 | Improved |
| add RNN temporal smoothing (ablation) | L3-Net · SanJoseDowntown | horizontal RMS: 0.057 -> 0.055 | Improved |
| add RNN temporal smoothing (ablation) | L3-Net · SunnyvaleBigLoop | horizontal RMS: 0.060 -> 0.055 | Improved |
| use mini-PointNet descriptors instead of PointNet (ablation) | L3-Net · Descriptor comparison | horizontal RMS: 0.130 -> 0.069 | Improved |
| use mini-PointNet descriptors instead of PPFNet (ablation) | L3-Net · Descriptor comparison | horizontal RMS: 0.085 -> 0.069 | Improved |
| remove 3D CNN cost-volume regularization (ablation) | L3-Net · 3D CNN configuration | horizontal RMS: 0.065 -> 0.134 | Hurt |
| reduce 3D CNN layers from 3 to 1 (ablation) | L3-Net · 3D CNN configuration | horizontal RMS: 0.065 -> 0.137 | Hurt |
| increase 3D CNN layers from 3 to 4 (ablation) | L3-Net · 3D CNN configuration | horizontal RMS: 0.065 -> 0.060 | Improved |
| use L3-Net rather than Wan et al. | LiDAR localization · SunnyvaleBigLoop | horizontal RMS: 0.069 -> 0.055 | Improved |
Closest papers by meaning
- LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving (CVPR 2019 · similarity 0.88)
- Pseudo-LiDAR From Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving (CVPR 2019 · similarity 0.88)
- DA4AD: End-to-End Deep Attention-based Visual Localization for Autonomous Driving (ECCV 2020 · similarity 0.87)
- LightLoc: Learning Outdoor LiDAR Localization at Light Speed (CVPR 2025 · similarity 0.85)
- An Empirical Study of the Generalization Ability of Lidar 3D Object Detectors to Unseen Domains (CVPR 2024 · similarity 0.85)
- DeepPointMap: Advancing LiDAR SLAM with Unified Neural Descriptors (AAAI 2024 · similarity 0.85)
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.
