Serviceeinschränkungen vom 12.-22.02.2026 - weitere Infos auf der UB-Homepage

Treffer: SSVM 2015: Scale Space and Variational Methods in Computer Vision / Artifact-free variational MPEG decompression

Title:
SSVM 2015: Scale Space and Variational Methods in Computer Vision / Artifact-free variational MPEG decompression
Authors:
Bredies, Kristian, Holler, Martin
Publication Year:
2015
Collection:
Graz University (UGR): Unipub
Subject Terms:
MPEG decompression, image sequence regularization, total generalized variation, inmal convolution type functionals
Subject Geographic:
UG:NW:MA
Document Type:
Buch book part<br />article in journal/newspaper
File Description:
text/html
Language:
English
ISBN:
978-3-319-18461-6
3-319-18461-X
Relation:
vignette : https://unipub.uni-graz.at/titlepage/urn/urn:nbn:at:at-ubg:3-5585/128; local:99144941083603331; system:AC15365474
Availability:
https://resolver.obvsg.at/urn:nbn:at:at-ubg:3-5585
Accession Number:
edsbas.28B1BED6
Database:
BASE

Weitere Informationen

We propose a variational method for artifact-free video de- compression that is capable of processing any MPEG-2 encoded movie. The method extracts, from a given MPEG-2 file, a set of admissible image sequences and minimizes an artifact-penalizing spatio-temporal regularization functional over this set, giving an optimal decompressed image sequence. For regularization, we use the infimal convolution of spatiotemporal Total Generalized Variation functionals (ICTGV). Numerical experiments on MPEG encoded files show that our approach significantly increases image quality compared to standard decompression.