Sequences of visual and haptic exploration were obtained on surfaces of different curvature from human subjects. We then extracted regions of interest (ROI) from the data as a function of number of times a subject fixated on a certain location on object and amount of time spent on such each location. Simple models like a plane, cone, cylinder, paraboloid, hyperboloid, ellipsoid, simple-saddle and a monkey-saddle were generated. Gaussian curvature representation of each point on all the surfaces was pre-computed. The surfaces have been previously tested for haptic and visual realism and distinctness by human subjects in a separate experiment. Both visual and haptic rendering were subsequently used for exploration by human subjects to study whether there is a similarity between the visual ROI and haptic ROIs. Additionally, we wanted to see if there is a correlation between curvature values and the ROIs thus obtained. A multiple regression model was further developed to see if this data can be used to predict the visual exploration path using haptic curvature saliency measures.

Original languageEnglish (US)
Title of host publicationFoundations of Augmented Cognition - Third International Conference, FAC 2007. Held as Part of HCI International 2007, Proceedings
PublisherSpringer Verlag
Number of pages8
ISBN (Print)9783540732150
StatePublished - 2007
Event3rd International Conference on Foundations of Augmented Cognition, FAC 2007 - Beijing, China
Duration: Jul 22 2007Jul 27 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4565 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other3rd International Conference on Foundations of Augmented Cognition, FAC 2007


  • Attention
  • Eye movements
  • Haptics
  • Regions of interest
  • Saliency
  • Vision

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)


Dive into the research topics of 'A model for visio-haptic attention for efficient resource allocation in multimodal environments'. Together they form a unique fingerprint.

Cite this