Iterative Grassmannian optimization for robust image alignment

作者：

Highlights：

• Propose an online algorithm t-GRASTA for the transformed robust PCA problem.

• Use a union of subspaces to approximate the nonlinear subspace learning process.

• Demonstrate the fully online mode of t-GRASTA with videos with camera jitter.

摘要

•Propose an online algorithm t-GRASTA for the transformed robust PCA problem.•Use a union of subspaces to approximate the nonlinear subspace learning process.•Demonstrate the fully online mode of t-GRASTA with videos with camera jitter.

论文关键词：Robust subspace learning,Grassmannian optimization,Image alignment,ADMM (alternating direction method of multipliers)

论文评审过程：Received 12 June 2013, Revised 27 November 2013, Accepted 21 February 2014, Available online 12 March 2014.

论文官网地址：https://doi.org/10.1016/j.imavis.2014.02.015