1 April 2008 Novel semisupervised high-dimensional correspondences learning method
Author Affiliations +
Correspondence is one of the big challenges in machine learning and image processing. To match two high-dimensional data sets with a certain number of aligned training examples, a novel semisupervised method is proposed. It is mainly based on two manifold learning approaches: maximum variance unfolding (MVU) and locally linear embedding (LLE). We have modified MVU to a semi-supervised version to solve the correspondence problem. Additionally, the nonuniform warps and folds caused by employing LLE alone and the computational burden of MVU disappear when they are combined. The proposed algorithm outperforms traditional methods in accuracy and efficiency. Three examples are performed to demonstrate the potential of this method.
© (2008) Society of Photo-Optical Instrumentation Engineers (SPIE)
Chenping Hou, Chenping Hou, Yi Wu, Yi Wu, Dongyun Yi, Dongyun Yi, Yuanyuan Jiao, Yuanyuan Jiao, "Novel semisupervised high-dimensional correspondences learning method," Optical Engineering 47(4), 047201 (1 April 2008). https://doi.org/10.1117/1.2903093 . Submission:


Real Time Visual Tracking Of Targets In Six Dimensions
Proceedings of SPIE (February 28 1990)
Integrated system for sensing and traverse of cliff faces
Proceedings of SPIE (September 29 2003)
Designing Vision Systems For Robotic Applications
Proceedings of SPIE (August 12 1989)
A clustering algorithm for intrusion detection
Proceedings of SPIE (March 27 2005)
Vision System That Learns How To Inspect Parts
Proceedings of SPIE (November 21 1982)

Back to Top