Abstract
In the human brain, the hippocampus performs a critical role in the formation of memory about the objects and places traced during real-world navigation. In a Virtual Reality (VR) system, an avatar needs to have the ability to remember the objects it once discovers. To inculcate the human-like remembering capability in an Intelligent Virtual Agent (IVA), this paper presents a novel model for Virtual Hippocampus (VH). Without using the contemporary data-hungry Machine Learning classifiers, objects are learnt and recognized on the basis of distinguishing features. Moreover, the processing time is not consumed in training of trivial data; rather an IVA learns new objects and places as unfolded to it during explicit navigation. The VH is made of two dynamic data structures; Distance with Direction (DD) and Vector of Positions with Areas (VPA). The data structures are updated as new routes are followed or unknown objects are discovered. The IVA makes the use of VH for self-directed navigation and automated selection of any known object. The model is implemented and evaluated in a case-study project; Automated Scene Learning (ASL). The satisfactory outcome (83%) of the evaluation assures applicability of the model in intelligent VR systems.
Presenters
Abdulrahman AlzahraniFaculty Member, Information Systems & Technology, University of Jeddah, Saudi Arabia
Details
Presentation Type
Paper Presentation in a Themed Session
Theme
2024 Special Focus—People, Education, and Technology for a Sustainable Future
KEYWORDS
Automation, Intelligent, Agents, Interactive, Systems, Learning, Virtual, Reality
Digital Media
This presenter hasn’t added media.
Request media and follow this presentation.