Utilizing gaze behavior for inferring task transitions using abstract hidden markov models

• We demonstrate an improved method for utilizing observed gaze behavior and show that it is useful in inferring hand movement intent during goal directed tasks. The task dynamics and the relationship between hand and gaze behavior are learned using an Abstract Hidden Markov Model (AHMM). We show that the predicted hand movement transitions occur consistently earlier in AHMM models with gaze than those models that do not include gaze observations.

Subject headings

NATURVETENSKAP  -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Computer Sciences (hsv//eng)

Keyword

AHM

Gaze Modeling

Gaze Tracking

Probabilistic Graphical Model

Task Transitions Prediction

Publication and Content Type

ref (subject category)

art (subject category)