Geometry-Aligned Semantic Matchingfor Cross-Modal Planar Registration

Zhiwei Wang1,2 Defeng He1,3 Yuxing Li2 Meilu Zhu2 Edmund Y. Lam2

1 College of Information Engineering, Zhejiang University of Technology, Hangzhou, 310023, China
2 Department of Electrical and Computer Engineering, The University of Hong Kong, Hong Kong, China
3 Provincial Key Laboratory of Intelligent Perception and Control for Complex Systems, Zhejiang University of Technology, Hangzhou, 310023, China
Paper Online Demo Code 🗂 Dataset
Teaser GIF 1 Teaser GIF 2

Generalization Performance

Although CDPM is not designed as a general cross-modal matcher, its design enables effective generalization across diverse cross-modal scenarios. CDPM online demo interface with image upload panels and cross-modal matching examples Try the Online Demo Generalization performance

Citation

@article{wang2026cdpm,
  title   = {Geometry-Aligned Semantic Matching for Cross-Modal Planar Registration},
  author  = {Wang, Zhiwei and He, Defeng and Li, Yuxing and Zhu, Meilu and Lam, Edmund Y.},
  journal = {xxxx},
  year    = {2026}
}