Fiche du document
16 juin 2017
- handle: 10670/1.aznbxs
- hal: hal-01656209
- doi: 10.1007/978-3-319-59105-6_15
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-319-59105-6_15
http://creativecommons.org/licenses/by/ , info:eu-repo/semantics/OpenAccess
Sujets proches
MudraCiter ce document
Łukasz Gadomer et al., « Gesture Recognition in 3D Space Using Dimensionally Reduced Set of Features », HAL-SHS : sciences de l'information, de la communication et des bibliothèques, ID : 10.1007/978-3-319-59105-6_15
Métriques
Partage / Export
Résumé
In this study authors present a solution to track and recognize arbitrary gestures of hands in three dimensional space and review the recognition accuracy. The idea of this novel gesture recognition system is described and results of research made on a recorded gesture data set are presented. Gesture instances were defined by user standing in different distances from the controller, in different placements of their field of vision and with different speeds, making recognition velocity and position invariant. Authors’ goal was to find the minimal number of features that give satisfying gesture classification result in order to achieve a compromise between accuracy and computation time. In this publication progress of the research on gesture recognition problem is described and a comparative study is presented.