An inexact splitting method for the subspace segmentation from incomplete and noisy observations

An inexact splitting method for the subspace segmentation from incomplete and noisy observations

Liang, R. (Shanghai University)
Bai, Y. (Shanghai University)
Lin, H.X. (TU Delft Mathematical Physics)

2018

Subspace segmentation is a fundamental issue in computer vision and machine learning, which segments a collection of high-dimensional data points into their respective low-dimensional subspaces. In this paper, we first propose a model for segmenting the data points from incomplete and noisy observations. Then, we develop an inexact splitting method for solving the resulted model. Moreover, we prove the global convergence of the proposed method. Finally, the inexact splitting method is implemented on the clustering problems in synthetic and benchmark data, respectively. Numerical results demonstrate that the proposed method is computationally efficient, robust as well as more accurate compared with the state-of-the-art algorithms.

Inexact augmented Lagrange multiplier method
Low rank representation
Subspace segmentation

http://resolver.tudelft.nl/uuid:4d9dd287-0744-43ad-a7aa-ca31046194c0

https://doi.org/10.1007/s10898-018-0684-4

2019-07-01

0925-5001

Journal of Global Optimization, 73 (2019), 411–429

Accepted author manuscript

Institutional Repository

journal article

© 2018 R. Liang, Y. Bai, H.X. Lin