Research map / Point clouds / Segmentation and detection
LWSIS: LiDAR-Guided Weakly Supervised Instance Segmentation for Autonomous Driving
Read the paperCheck 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 |
|---|---|---|---|
| Add LWSIS to PointSup | PointSup with ResNet50 · nuInsSeg val | AP: 43.80 -> 45.46 | Helped |
| Add LWSIS to PointSup | PointSup with ResNet101 · nuInsSeg val | AP: 44.72 -> 46.17 | Helped |
| Add LWSIS to BoxInst | BoxInst with ResNet50 · nuInsSeg val | AP: 33.40 -> 35.65 | Helped |
| Add LWSIS to BoxInst | BoxInst with ResNet101 · nuInsSeg val | AP: 34.18 -> 36.22 | Helped |
| Add LWSIS to BoxInst | BoxInst · Waymo val | AP: 34.53 -> 37.77 | Helped |
| Use LWSIS segmentation labels in PointPainting | CenterPoint · nuScenes val | mAP: 56.7 -> 62.2 | Helped |
| Add PLA supervision (ablation) | BoxInst · nuInsSeg val | AP: 33.40 -> 34.77 | Helped |
| Add image-similarity GCR (ablation) | BoxInst · nuInsSeg val | AP: 33.40 -> 34.32 | Helped |
| Add geometry-similarity GCR (ablation) | BoxInst · nuInsSeg val | AP: 33.40 -> 34.15 | Helped |
| Add image-similarity GCR alongside PLA (ablation) | BoxInst with PLA · nuInsSeg val | AP: 34.77 -> 35.27 | Helped |
| Add geometry-similarity GCR to PLA and image GCR (ablation) | BoxInst with PLA and image GCR · nuInsSeg val | AP: 35.27 -> 35.65 | Helped |
| Add PLA supervision (ablation) | BoxInst · nuInsSeg val | AP_l: 50.21 -> 55.43 | Helped |
Closest papers by meaning
- MWSIS: Multimodal Weakly Supervised Instance Segmentation with 2D Box Annotations for Autonomous Driving (AAAI 2024 · similarity 0.92)
- Weakly Supervised Semantic Segmentation for Driving Scenes (AAAI 2024 · similarity 0.89)
- Exploring Scene Affinity for Semi-Supervised LiDAR Semantic Segmentation (CVPR 2025 · similarity 0.88)
- VLM2Scene: Self-Supervised Image-Text-LiDAR Learning with Foundation Models for Autonomous Driving Scene Understanding (AAAI 2024 · similarity 0.88)
- HiLoTs: High-Low Temporal Sensitive Representation Learning for Semi-Supervised LiDAR Segmentation in Autonomous Driving (CVPR 2025 · similarity 0.88)
- Multi-Space Alignments Towards Universal LiDAR Segmentation (CVPR 2024 · similarity 0.87)
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.