Research map / Detection and segmentation / 3D detection and driving
Unsupervised 3D Perception with 2D Vision-Language Distillation for Autonomous Driving
Mahyar Najibi, Jingwei Ji, Yin Zhou, Charles R. Qi, Xinchen Yan, Scott Ettinger, Dragomir Anguelov
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 vision-language feature distillation head to the 3D detector (ablation) | UP-VL detector · WOD class-aware all motion states | mAP: 27.6 -> 35.8 | Improved |
| train detector on auto labels instead of using auto labels directly | UP-VL · WOD class-aware moving objects | mAP: 43.7 -> 52.8 | Improved |
| use amodal boxes and VL features vs visible-box clustering baseline | auto labeling · WOD class-aware all motion states | mAP: 13.1 -> 22.4 | Improved |
| add vision-language features to clustering (ablation) | clustering baseline · WOD class-aware all motion states | mAP: 11.6 -> 12.9 | Improved |
| add scene-flow motion features to clustering with VL features (ablation) | clustering baseline · WOD class-aware all motion states | mAP: 12.9 -> 13.1 | No clear effect |
| raise background-ratio filtering threshold r_bg from 50% to 99% (ablation) | auto labeling · WOD all motion states auto labels | mAP: 14.1 -> 22.4 | Improved |
| change background cosine threshold from 0.02 to 0.10 (ablation) | auto labeling · WOD all motion states auto labels | mAP: 22.4 -> 20.5 | Hurt |
| use UP-VL instead of MI-UP for class-agnostic moving-object detection | unsupervised 3D detector · WOD class-agnostic moving objects, IoU 0.5 L1 | 3D AP: 29.6 -> 38.4 | Improved |
Closest papers by meaning
- 3D Annotation-Free Learning by Distilling 2D Open-Vocabulary Segmentation Models for Autonomous Driving (AAAI 2025 · similarity 0.89)
- UniPAD: A Universal Pre-training Paradigm for Autonomous Driving (CVPR 2024 · similarity 0.87)
- VLM2Scene: Self-Supervised Image-Text-LiDAR Learning with Foundation Models for Autonomous Driving Scene Understanding (AAAI 2024 · similarity 0.87)
- OmniDrive: A Holistic Vision-Language Dataset for Autonomous Driving with Counterfactual Reasoning (CVPR 2025 · similarity 0.86)
- MAPLM: A Real-World Large-Scale Vision-Language Benchmark for Map and Traffic Scene Understanding (CVPR 2024 · similarity 0.86)
- OpenAD: Open-World Autonomous Driving Benchmark for 3D Object Detection (NeurIPS 2025 · 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.
