Research map / Point clouds / Segmentation and detection
Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Driving
Lingdong Kong, Xiang Xu, Jiawei Ren, Wenwei Zhang, Liang Pan, Kai Chen, Wei Tsang Ooi, Ziwei Liu
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 |
|---|---|---|---|
| add camera-to-LiDAR distillation and language guidance on top of LaserMix | LaserMix with FIDNet · nuScenes (1% labels) | mIoU: 49.5 -> 51.6 | Improved |
| add multi-modal components to LaserMix | LaserMix with Cylinder3D · SemanticKITTI (1% labels) | mIoU: 50.6 -> 56.2 | Improved |
| add multi-modal components to LaserMix | LaserMix with SPVCNN · ScribbleKITTI (50% labels) | mIoU: 61.0 -> 66.2 | Improved |
| add multi-modal components to LaserMix | LaserMix with MinkUNet · nuScenes (1% labels) | mIoU: 62.8 -> 64.7 | Improved |
| add multi-modal components to LaserMix | LaserMix with Cylinder3D · ScribbleKITTI (10% labels) | mIoU: 60.3 -> 56.7 | Hurt |
| add multi-modal components to LaserMix | LaserMix with Cylinder3D · ScribbleKITTI (20% labels) | mIoU: 55.1 -> 57.6 | Hurt |
| use LaserMix++ over LaserMix | MinkUNet · SemanticKITTI-C | mCE: 83.3 -> 80.9 | Improved |
| use LaserMix++ over LaserMix | CenterPoint · WOD-C | mCE: 100.5 -> 98.2 | Improved |
| use LaserMix++ over LaserMix | MinkUNet · SemanticKITTI-C | mRR: 87.0 -> 88.1 | No clear effect |
| add LaserMix++ as pretraining-plus-semi-supervised step instead of LaserMix | Seal-pretrained backbone · nuScenes (1% labels) | mIoU: 48.4 -> 49.9 | Improved |
| add LaserMix++ over LaserMix | Seal-pretrained backbone · nuScenes (full labels) | mIoU: 77.1 -> 77.9 | Improved |
| add LaserMix++ over LaserMix | FIDNet · nuScenes (50% labels) | mIoU: 73.0 -> 73.7 | Improved |
Closest papers by meaning
- Temporal Consistent 3D LiDAR Representation Learning for Semantic Perception in Autonomous Driving (CVPR 2023 · similarity 0.90)
- Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data (CVPR 2022 · similarity 0.90)
- VLM2Scene: Self-Supervised Image-Text-LiDAR Learning with Foundation Models for Autonomous Driving Scene Understanding (AAAI 2024 · similarity 0.89)
- HiLoTs: High-Low Temporal Sensitive Representation Learning for Semi-Supervised LiDAR Segmentation in Autonomous Driving (CVPR 2025 · similarity 0.88)
- Exploring Scene Affinity for Semi-Supervised LiDAR Semantic Segmentation (CVPR 2025 · similarity 0.88)
- Multi-Space Alignments Towards Universal LiDAR Segmentation (CVPR 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.
