Subspace Learning Research

My research in problem formulations based on subspace learning concerns its evaluation and its applications to background/foreground separation and background initialization. Thus, my main research on subspace learning can be summarized in the following points.

1. Surveys

• A Survey Approach: My surveys concern the application of reconstructive subspace learning in background subtraction as well as the application of robust subspace learning (RPCA via L+S decomposition) in background/foreground separation. (2 Chapters, 4 Journals, 1 Conference)

2. Regular Subspace Learning

• Discriminative Approach: This research concerns the evaluation of a discriminative subspace learning called Incremental Maximum Margin Criterion (IMMC) for background subtraction. (1 Journal, 2 Conferences)
• Mixed Approach: This research concerns the use of both PCA and LDA for background modeling and foreground detection, respectively. (1 Conference)

3. Robust Subspace Learning/Dynamic Subspace Learning

• Robust Matrix Approach: This research concerns RPCA via L+S decomposition into matrices for background/foreground separation and background initialization. (3 Chapters, 7 Journals, 18 Conferences)
• Robust Tensor Approach: This research concerns RPCA via L+S decomposition into tensors for background/foreground separation and background initialization. (4 Conferences)

Publications (37) : 3 (5-2) Chapters, 9 (12-3) Journals, 25 (26-1) Conferences

Note: My publications are available on Academia, Research Gate, Researchr, ORCID and Publication List.