• English
    • Deutsch
  • English 
    • English
    • Deutsch
  • Login
View Item 
  •   Home
  • Universität Ulm
  • Publikationen
  • View Item
  •   Home
  • Universität Ulm
  • Publikationen
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

New methods for anechoic demixing with application to shift invariant feature extraction

Thumbnail
vts_7243_10225.pdf (62.56Mb)
128 Seiten
Veröffentlichung
2010-03-26
Authors
Omlor, Lars
Dissertation


Faculties
Fakultät für Ingenieurwissenschaften und Informatik
Abstract
Blind source separation problems emerge in many applications, where signals can be modeled as superpositions of multiple sources. Many popular applications of blind source separation are based on linear instantaneous mixture models. If specific invariance properties are known about the sources, e.g. translation or rotation invariance, the simple linear model can be extended by inclusion of the corresponding transformations. When the sources are invariant against translations (i.e. spatial displacements or time shifts) the resulting model is called anechoic mixing model. The main focus of this thesis is the development of new mathematical framework for the solution of the anechoic mixing problem and the successive derivation of concrete algorithms. This framework integrates approaches from many distinct fields of signal processing like stochastic time-frequency analysis, convex optimization, projection onto convex set methods, delay estimation and naturally blind source separation. The developed method is tested on a variety of applications including music recordings, natural two dimensional images, two-dimensional shapes and optic flow. However the main application is the analysis and synthesis of human motion trajectories, which is motivated by the idea in motor control that complex motor behavior can be explained by a superposition of simple basis components, or spatio-temporal primitives. The new anechoic demixing algorithm allows to approximate high-dimensional movement trajectories accurately based on a small number of learned primitives or source signals. It is demonstrated that the new method is significantly more accurate than other common techniques. This allows the modeling of subtle style changes, like the bodily expression of emotion as well as a sufficient synthesis quality for computer animation with only few mixture components.
Date created
2010
Subject headings
[GND]: Maschinelles Lernen | Zeit-Frequenz-Analyse
[LCSH]: Blind source separation
[MeSH]: Man-machine systems
[Free subject headings]: Anechoic mixing | Independent component analysis | Time-frequency analysis
[DDC subject group]: DDC 004 / Data processing & computer science
License
Standard (Fassung vom 01.10.2008)
https://oparu.uni-ulm.de/xmlui/license_v2

Metadata
Show full item record

DOI & citation

Please use this identifier to cite or link to this item: http://dx.doi.org/10.18725/OPARU-1722

Omlor, Lars (2010): New methods for anechoic demixing with application to shift invariant feature extraction. Open Access Repositorium der Universität Ulm und Technischen Hochschule Ulm. Dissertation. http://dx.doi.org/10.18725/OPARU-1722
Citation formatter >



Policy | kiz service OPARU | Contact Us
Impressum | Privacy statement
 

 

Advanced Search

Browse

All of OPARUCommunities & CollectionsPersonsInstitutionsPublication typesUlm SerialsDewey Decimal ClassesEU projects UlmDFG projects UlmOther projects Ulm

My Account

LoginRegister

Statistics

View Usage Statistics

Policy | kiz service OPARU | Contact Us
Impressum | Privacy statement