7th Annual Meeting of the International Multisensory Research Forum
    Home > Papers > Yasuyuki Inoue
Yasuyuki Inoue

Recognition of Human Body Movements: View-dependency, Inverse effect and biomechanical constraints.
Poster Presentation

Yasuyuki Inoue
Department of Knowledge-based Information Engineering, Toyohashi University of Technology

Yasuhisa Fujiki
Department of Knowledge-based Information Engineering, Toyohashi University of Technology

Michiteru Kitazaki
Research Center for Future Vehicle / Intelligent Sensing System Research Center, Toyohashi University of Technology

     Abstract ID Number: 131
     Full text: Not available
     Last modified: March 18, 2006
     Presentation date: 06/20/2006 10:00 AM in Hamilton Building, Foyer
     (View Schedule)

Abstract
Human body is represented multimodally as visual and motor in the brain. We reported that visual recognition of static body posture was affected by biomechanical constraints (Kitazaki and Inoue, VSS 2004), and action of the observer modulated the visual recognition of body posture (Inoue and Kitazaki, IMRF 2005). In this study, we investigated the effects of biomechanical constraints of human body on visual recognition of body movements, view-dependency and its inversion effects.
First, we made twenty-seven different body movements with a three-dimensional computer modeling software. They were all biomechanically possible movements. Then, twenty-seven impossible movements were made by modifying possible ones. A pair of movements across viewpoints (0-180deg) was presented sequentially, and subjects were asked if they were identical or different movements regardless of the view difference. We found that recognition performance of possible movements was better and more view-independent than that of impossible ones. These effects decreased with using inverted bodies.
Thus, recognition of human body movements was affected by biomechanical relationship of body parts and movements, and it was specific to upright body representation. These results suggest that the brain processing for human-body recognition is multimodal and utilizes both visual and motor information.

[Supported by Nissan-Science-Foundation and MEXT-Japan]

Research
Support Tool
  For this 
refereed conference abstract
Capture Cite
View Metadata
Printer Friendly
Context
Author Bio
Define Terms
Related Studies
Media Reports
Google Search
Action
Email Author
Email Others
Add to Portfolio



    Learn more
    about this
    publishing
    project...


Public Knowledge

 
Open Access Research
home | overview | program
papers | organization | schedule | links
  Top