Grassmann Manifold Optimization for Fast $L_1$-Norm Principal Component Analysis

In this letter, we propose a fast Grassmann manifold optimization method for L
₁
-norm based principal component analysis (GM-L
₁
-PCA). Our approach is a two-step iterative costminimization and manifold retraction technique that efficiently finds all principal components simultaneously. We perform complexity analysis and show that GM-L
₁
-PCA achieves a significant reduction in processing time while obtaining comparable or better results to current state-of-the-art L
₁
-PCA methods. We further demonstrate the improvement of GM-L
₁
-PCA technique over L
₂
-PCA on a dataset of facial imagery corrupted with outlying data points. Our experiments show that GM-L
₁
-PCA is computationally more efficient and produces results with lower reprojection error than previous methods. Furthermore, the processing time of our approach is relatively independent of dataset size and well suited for various big-data problems commonly encountered today.