Zur Seitenansicht
 

Titelaufnahme

Titel
Contour-Guided Gaze Gestures: Eye-based Interaction with Everyday Objects / submitted by Florian Jungwirth, BSc
AutorInnenJungwirth, Florian
Beurteiler / BeurteilerinFerscha, Alois
ErschienenLinz, 2017
Umfang77 Seiten : Illustrationen
HochschulschriftUniversität Linz, Masterarbeit, 2017
SpracheEnglisch
DokumenttypMasterarbeit
Schlagwörter (EN)eye-tracking / pervasive computing / human-computer interaction / gaze-based interaction
Schlagwörter (GND)Elektrookulographie / Ubiquitous Computing / Mensch-Maschine-Kommunikation / Blickbewegung
URNurn:nbn:at:at-ubl:1-19239 Persistent Identifier (URN)
Zugriffsbeschränkung
 Das Werk ist gemäß den "Hinweisen für BenützerInnen" verfügbar
Dateien
Contour-Guided Gaze Gestures: Eye-based Interaction with Everyday Objects [14.86 mb]
Links
Nachweis
Klassifikation
Zusammenfassung (Englisch)

The eyes are gaining increasing interest within the HCI (human-computer interaction) community, as they are a fast and accurate input modality for interactive systems. However, the applicability of mobile eye-based HCI so far is restricted by several issues, such as calibration or the Midas Touch Problem [Jac91]. In this thesis, I propose the idea of contour-guided gaze gestures, which overcome these problems by relying on relative eye movements, as users trace the contours of (interactive) objects within a smart environment. Matching the trajectory of the eye movements and the contour shapes allows to estimate which object was interacted with and to trigger the corresponding actions. The interaction concept and the design of the system are described, along with the influence of several parameters on the gesture detection performance, that was evaluated in a first user study. Using these findings, a second study with three application scenarios was executed and demonstrated, that users were able to trace object contours to trigger actions from various positions on multiple different objects. It was further determined, that the proposed method is an easy to learn, hands-free interaction technique, that is robust against false positive activations. User study results highlight low demand values and show that the method holds potential for further exploration, but also reveal areas for refinement.

Statistik
Das PDF-Dokument wurde 30 mal heruntergeladen.