A splitting algorithm for directional regularization and sparsification

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

Lars Lau Rakêt, Mads Nielsen

We present a new split-type algorithm for the minimization
of a p-harmonic energy with added data fidelity
term. The half-quadratic splitting reduces the
original problem to two straightforward problems, that
can be minimized efficiently. The minimizers to the two
sub-problems can typically be computed pointwise and
are easily implemented on massively parallel processors.
Furthermore the splitting method allows for the
computation of solutions to a large number of more advanced
directional regularization problems. In particular
we are able to handle robust, non-convex data terms,
and to define a 0-harmonic regularization energy where
we sparsify directions by means of an L0 norm.
OriginalsprogEngelsk
TitelProceedings of the 21st International Conference on Pattern Recognition (ICPR)
Antal sider5
ForlagIEEE
Publikationsdato2012
Sider3094-3098
ISBN (Trykt)978-4-9906441-0-9
StatusUdgivet - 2012
BegivenhedInternational Conference on Pattern Recognition - Tsukuba International Congress Center, Tsukuba, Japan
Varighed: 11 nov. 201215 nov. 2012
Konferencens nummer: 21

Konference

KonferenceInternational Conference on Pattern Recognition
Nummer21
LokationTsukuba International Congress Center
LandJapan
ByTsukuba
Periode11/11/201215/11/2012

ID: 38468530