Riemannian Robust Principle Component Pursuit (R2PCP)
http://www.uni-graz.at/imawww/ifb/r2pcp/index.html
________________________________________________________________
OVERVIEW
Robust principle component pursuit (RPCP) aims at decomposing a given data
matrix additively into a low-rank matrix and a sparse matrix. R2PCP
solves such a matrix decomposition problem by formulating a least-
squares problem subject to rank- and cardinality constraints. The
resulting constrained optimization problem is tackled by an (inexact)
alternating minimization scheme on the respective matrix manifolds.
In particular, the low-rank matrix subproblem is resolved by a
Riemannian optimization step. In addition, based on particular
applications, a heuristic trimming procedure is incorporated to adjust
the rank and cardinality dynamically.
The R2PCP package was implemented in MATLAB.
_________________________________________________________________
REFERENCE
Michael Hintermueller and Tao Wu, "Robust Principal Component Pursuit
via Inexact Alternating Minimization on Matrix Manifolds", Journal
of Mathematical Imaging and Vision, to appear.
DOI:10.1007/s10851-014-0527-y
_________________________________________________________________
DIRECTORY STRUCTURE
demo_synthetic_data.m
Demonstration of R2PCP on synthetic data.
demo_airport_video.m
Demonstration of R2PCP for background-foreground separation
on the "airport" video.
demo_lobby_video.m
Demonstration of R2PCP for background-foreground separation
on the "lobby" video.
README.txt
This file.
Riem_RPCP (dir)
Source codes of R2PCP.
video_lobby (dir)
Raw data for the "airport" video.
video_lobby (dir)
Raw data for the "lobby" video.
utilities (dir)
Utility functions used in demonstrations.
_________________________________________________________________
ACKNOWLEDGEMENT
The R2PCP package was developed by Michael Hintermueller from
Department of Mathematics at the Humboldt-University of Berlin and
Tao Wu from Institute for Mathematics and Scientific Computing at
the University of Graz. This work is supported by the Austria
Science Fund (FWF) under START-program Y305 "Interfaces and Free
Boundaries" and the SFB F32 "Mathematical Optimization and
Applications in Biomedical Sciences".
_________________________________________________________________
DISCLAIMER
The R2PCP package (including code modifications) may only be used
for NON-COMMERCIAL RESEARCH purposes. For inquiries concerning a
different use, please contact Prof. Michael Hintermueller from the
Humboldt-University of Berlin (hint"at"math.hu-berlin.de).
Your comments are welcome. Please keep track of bugs or missing/
confusing instructions and report them to
Michael Hintermueller
Tao Wu
The algorithms contained in the R2PCP package were implemented by
Tao Wu.