Abstract

AbstractUser generated virtual worlds, such as Second Life, typically lack accurate metadata for their virtual world objects. This is a significant problem for blind users who rely on textual descriptions in order to access virtual worlds using synthetic speech. In this paper, we consider the problem of automatic object labeling to improve accessibility of virtual worlds for users with disabilities. Taking advantage of the primitivebased representation of virtual world objects in Second Life, we present an approach that leverages histogram-based geometric object representations, machine learning and crowdsourcing to accurately label virtual world objects at a large scale. We report excellent classification results using seven challenging object classes.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.