arXiv：cs.AI· Jiawen Yang, Shuhao Chen, Shengtao Zhang, Ke Tang, Yu Zhang·· 3 小时前
通过潜在空间桥接实现异构模态无监督域适应
Heterogeneous-Modal Unsupervised Domain Adaptation via Latent Space Bridging
AI 导读
研究者提出异构模态无监督域适应（HMUDA）新设定，通过一个包含两种模态配对观测的无标注桥接域实现跨模态知识迁移，桥接域分布可偏离源域和目标域。对应方法 Latent Space Bridging（LSB）采用双分支框架，以配对桥接样本上的特征一致性损失弥合模态差距，并用类质心对齐损失缩小源目标差异。在覆盖 2D 到 3D 与 3D 到 2D 迁移的八个基准设定上，LSB 取得了 SOTA 表现。
正文
Abstract:Unsupervised domain adaptation (UDA) effectively bridges the domain gap between a labeled source domain and an unlabeled target domain, but assumes that the two domains share the same modality. Heterogeneous domain adaptation (HDA) instead handles different feature spaces across domains, yet requires labeled target samples or paired data linking the source and target domains. Neither applies when a labeled source domain and a fully unlabeled target domain each hold an entirely distinct modality (e.g., 2D images and 3D point clouds). To address this limitation, we introduce a new setting termed Heterogeneous-Modal Unsupervised Domain Adaptation (HMUDA), which transfers knowledge across modalities via an unlabeled bridge domain containing paired observations from both modalities, whose distribution may deviate from those of the source and target domains. To learn under the HMUDA setting, we propose Latent Space Bridging (LSB), a dual-branch framework where a feature consistency loss on paired bridge samples closes the modality gap and a class-centroid alignment loss reduces the source-target discrepancy. Extensive experiments on eight benchmark settings covering both 2D-to-3D and 3D-to-2D transfer demonstrate that LSB achieves state-of-the-art performance.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2506.15971 [cs.CV] |
| (or arXiv:2506.15971v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2506.15971 arXiv-issued DOI via DataCite |
Submission history
From: Jiawen Yang [view email]
[v1]
Thu, 19 Jun 2025 02:31:51 UTC (3,652 KB)
[v2]
Thu, 8 Oct 2026 08:37:39 UTC (5,262 KB)
来源：arXiv：cs.AI · arxiv.org