Logo des Repositoriums
Zur Startseite
  • English
  • Deutsch
Anmelden
  1. Startseite
  2. SuUB
  3. Dissertationen
  4. Model-Based High-Dimensional Pose Estimation with Application to Hand Tracking
 
Zitierlink URN
https://nbn-resolving.de/urn:nbn:de:gbv:46-00102865-17

Model-Based High-Dimensional Pose Estimation with Application to Hand Tracking

Veröffentlichungsdatum
2012-10-12
Autoren
Mohr, Daniel  
Betreuer
Zachmann, Gabriel  
Gutachter
Klinker, Gudrun  
Zusammenfassung
This thesis presents novel techniques for computer vision based full-DOF human hand motion estimation. Our main contributions are: A robust skin color estimation approach; A novel resolution-independent and memory efficient representation of hand pose silhouettes, which allows us to compute area-based similarity measures in near-constant time; A set of new segmentation-based similarity measures; A new class of similarity measures that work for nearly arbitrary input modalities; A novel edge-based similarity measure that avoids any problematic thresholding or discretizations and can be computed very efficiently in Fourier space; A template hierarchy to minimize the number of similarity computations needed for finding the most likely hand pose observed; And finally, a novel image space search method, which we naturally combine with our hierarchy. Consequently, matching can efficiently be formulated as a simultaneous template tree traversal and function maximization.
Schlagwörter
Computer Vision

; 

Object Detection

; 

Object Recognition

; 

Tracking

; 

Hand Pose Estimation
Institution
Universität Bremen  
Fachbereich
Fachbereich 03: Mathematik/Informatik (FB 03)  
Dokumenttyp
Dissertation
Zweitveröffentlichung
Nein
Sprache
Englisch
Dateien
Lade...
Vorschaubild
Name

00102865-1.pdf

Size

24.04 MB

Format

Adobe PDF

Checksum

(MD5):f1f23abbd0ee104fb9b3ff86a5ae7db6

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science

  • Datenschutzbestimmungen
  • Endnutzervereinbarung
  • Feedback schicken