Research map / Detection and segmentation / 3D detection and driving
OWL: Unsupervised 3D Object Detection by Occupancy Guided Warm-up and Large Model Priors Reasoning
Xusheng Guo, Wanfa Zhang, Shijia Zhao, Qiming Xia, Xiaolong Xie, Mingming Wang, Hai Wu, Chenglu Wen
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 Instance-Cued Reasoning (LLM-based pseudo-label filtering and refinement) (ablation) | OWL · WOD validation | 3D AP L1: 19.59 -> 30.25 | Improved |
| add Occupancy Guided Warm-up (ablation) | OWL (PLI+ICR) · WOD validation | 3D AP L1: 30.25 -> 34.16 | Improved |
| add Weight-adapted Self-training (ablation) | OWL (PLI+ICR+OGW) · WOD validation | 3D AP L1: 34.16 -> 36.67 | Improved |
| use contrastive learning warm-up instead of no warm-up (ablation) | OWL · WOD validation | AP L1: 30.25 -> 29.10 | Hurt |
| use point completion warm-up instead of no warm-up (ablation) | OWL · WOD validation | AP L1: 30.25 -> 31.06 | Improved |
| use OGW warm-up instead of no warm-up (ablation) | OWL · WOD validation | AP L1: 30.25 -> 34.16 | Improved |
| OWL vs CPD on unsupervised mAP | OWL · WOD validation | mAP L1: 19.59 -> 36.91 | Improved |
| OWL vs CPD at long range | OWL · WOD validation, [50,inf) m | 3D mAP: 5.14 -> 13.56 | Improved |
| OWL (unsupervised) vs CoIn weakly supervised on KITTI moderate | OWL · KITTI val Car AP@0.7 Moderate | 3D AP: 75.32 (CoIn) vs 62.67 (OWL) | Hurt |
Closest papers by meaning
- UNION: Unsupervised 3D Object Detection using Object Appearance-based Pseudo-Classes (NeurIPS 2024 · similarity 0.88)
- CorrBEV: Multi-View 3D Object Detection by Correlation Learning with Multi-modal Prototypes (CVPR 2025 · similarity 0.87)
- SOGDet: Semantic-Occupancy Guided Multi-View 3D Object Detection (AAAI 2024 · similarity 0.87)
- UnO: Unsupervised Occupancy Fields for Perception and Forecasting (CVPR 2024 · similarity 0.86)
- Achieving Speed-Accuracy Balance in Vision-based 3D Occupancy Prediction via Geometric-Semantic Disentanglement (AAAI 2025 · similarity 0.86)
- Training an Open-Vocabulary Monocular 3D Detection Model without 3D Data (NeurIPS 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.
