Research map / Point clouds / Segmentation and detection
Cross-Modal Contrastive Learning for Domain Adaptation in 3D Semantic Segmentation
Bowei Xing, Xianghua Ying, Ruibin Wang, Jinfa Yang, Taiyan Chen
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 cross-modal contrastive learning and neighborhood feature aggregation (ablation) | xMUDA · USA → Singapore | mIoU (2D+3D): 62.7 -> 64.5 | Improved |
| Add neighborhood feature aggregation module (ablation) | xMUDA · USA → Singapore | mIoU (2D+3D): 62.7 -> 63.8 | Improved |
| Add cross-modal contrastive loss (ablation) | xMUDA · USA → Singapore | mIoU (2D+3D): 62.7 -> 63.0 | Improved |
| Add pseudo-labeling (ablation) | Ours · USA → Singapore | mIoU (2D+3D): 64.5 -> 66.7 | Improved |
| Add pseudo-labeling | Ours · Day → Night | mIoU (2D+3D): 51.6 -> 54.5 | Improved |
| Add pseudo-labeling | Ours · A2D2 → SemanticKITTI | mIoU (2D+3D): 48.0 -> 52.8 | Improved |
| Add dilated-neighbor attention to NFAM (ablation) | NFAM · USA → Singapore | mIoU (2D+3D): 63.6 -> 63.8 | Improved |
| Use the proposed cross-modal contrastive learning and NFAM | xMUDA · Day → Night | mIoU (2D+3D): 50.0 -> 51.6 | Improved |
| Use the proposed cross-modal contrastive learning and NFAM | xMUDA · A2D2 → SemanticKITTI | mIoU (2D+3D): 44.0 -> 48.0 | Improved |
Closest papers by meaning
- Cross-Modal Learning for Domain Adaptation in 3D Semantic Segmentation (TPAMI 2022 · similarity 0.96)
- Sparse-to-Dense Feature Matching: Intra and Inter Domain Cross-Modal Learning in Domain Adaptation for 3D Semantic Segmentation (ICCV 2021 · similarity 0.94)
- Omni-Query Active Learning for Source-Free Domain Adaptive Cross-Modality 3D Semantic Segmentation (AAAI 2025 · similarity 0.88)
- UniDSeg: Unified Cross-Domain 3D Semantic Segmentation via Visual Foundation Models Prior (NeurIPS 2024 · similarity 0.88)
- EAGLE: Efficient Adaptive Geometry-based Learning in Cross-view Understanding (NeurIPS 2024 · similarity 0.87)
- Open-Set Domain Adaptation for Semantic 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.
