Research map / Label-efficient and robust learning / Domain adaptation and generalization
UniDxMD: Towards Unified Representation for Cross-Modal Unsupervised Domain Adaptation in 3D Semantic Segmentation
Zhengyin Liang, Hui Yin, Min Liang, Qianqian Du, Ying Yang, Hua Huang
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 A-VQ quantizer (ablation) | xMUDA · nuScenes: USA/Sing. | xM mIoU: 69.4 -> 69.9 | No clear effect |
| add CSQM (ablation) | xMUDA · nuScenes: USA/Sing. | xM mIoU: 69.4 -> 72.3 | Improved |
| add LSR to A-VQ (ablation) | xMUDA + A-VQ · nuScenes: USA/Sing. | xM mIoU: 69.9 -> 71.9 | Improved |
| add LSR to CSQM (ablation) | UniDxMD · nuScenes: USA/Sing. | xM mIoU: 72.3 -> 74.3 | Improved |
| use joint prototypes instead of source 2D prototypes only (ablation) | UniDxMD · nuScenes: USA/Sing. | xM mIoU: 73.0 -> 74.3 | Improved |
| use joint prototypes instead of target 3D prototypes only (ablation) | UniDxMD · nuScenes: USA/Sing. | xM mIoU: 72.4 -> 74.3 | Improved |
| add joint relation constraint loss (ablation) | UniDxMD · nuScenes: Day/Night | xM mIoU: 73.3 -> 74.6 | Improved |
| add joint relation constraint loss (ablation) | UniDxMD · nuScenes: USA/Sing. | xM mIoU: 73.1 -> 74.3 | Improved |
| replace 2D encoder with frozen CLIP ViT-Large | UniDxMD · v.KITTI/Sem.KITTI | xM mIoU: 58.7 -> 63.3 | Improved |
| replace 2D encoder with frozen CLIP ViT-Large | UniDxMD · A2D2/Sem.KITTI | xM mIoU: 52.5 -> 57.0 | Improved |
| compare UniDxMD against xMUDA | UniDxMD vs xMUDA · v.KITTI/Sem.KITTI | xM mIoU: 48.2 -> 58.7 | Improved |
| vary codebook size K and dimension D (ablation) | UniDxMD · nuScenes: USA/Sing. | xM mIoU | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- xMUDA — xMUDA: Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation (CVPR 2020) · better on 4 of 4 shared results
- UniDSeg — UniDSeg: Unified Cross-Domain 3D Semantic Segmentation via Visual Foundation Models Prior (NeurIPS 2024) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- xMUDA: Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation (CVPR 2020 · similarity 0.93)
- UniDSeg: Unified Cross-Domain 3D Semantic Segmentation via Visual Foundation Models Prior (NeurIPS 2024 · similarity 0.93)
- Cross-Modal Learning for Domain Adaptation in 3D Semantic Segmentation (TPAMI 2022 · similarity 0.92)
- Cross-Modal Contrastive Learning for Domain Adaptation in 3D Semantic Segmentation (AAAI 2023 · similarity 0.91)
- Universal Domain Adaptation for Semantic Segmentation (CVPR 2025 · similarity 0.89)
- Towards Unsupervised Domain Bridging via Image Degradation in Semantic Segmentation (NeurIPS 2025 · similarity 0.89)
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.
