|
home
teaching
research interests
current projects
portfolio
students
thesis topics
work opportunities
publications
resources
private
|
 | |
3D Display and Capture of Humans for Live-Virtual Training
- web: project web page
- description: We are carrying out research, development, and evaluation of work in both 3D display and sensing, initially aimed at Mixed/Augmented Reality-based live-virtual training, but with potential for broader impact. The first component of the work is aimed at a new idea for creating physically present virtual humans in 3D, without the need for HMDs or special time/phase/wavelength glasses. Our basic approach is to extend our earlier work in projector-based Shader Lamps to dynamically remap the shape and appearance of virtual or real humans onto human-shaped display surfaces. The second component of the work is aimed at a new idea for hybrid global/body-relative and cooperative estimation of full kinematic body and weapon posture, supporting multiple individuals working in close proximity during training exercises, for both kinetic and non-kinetic scenarios. The third component of the work is aimed at obtaining a thorough understanding of a relevant training domain, its actors, and their specific needs (both current and projected). This involves the investigation and reporting on the basic elements of their acting in related environment in the context of the 3D display and sensing efforts.
MOVES segment: domain analysis of three different training environments/domains (fire drills on ships and submarines, urban warfare operations, tactical cultural awareness training), and a series of user studies in a domain of Shader Lamps-based physical avatars.
- sponsor: Office of Naval Research (ONR)
- collaborating institutions: University of North Carolina at Chapel Hill (UNC) - lead, and Naval Postgraduate School (NPS)
- my role: co-PI on joint project (FY09 - FY12: four years project)
|
|
 | |
Behavioral Analysis and Synthesis for Intelligent Training - BASE-IT
- web: project web page ,
concept figure
- description: A primary objective is to develop a state-of-the-art intelligent training system for pre/in/post evaluation using behavior analysis, review, and behavior synthesis as well as a set of training approaches, aimed to support a wide range of training and operational needs in preparing for and conducting the training in MOUT facilities.
Our proposed approach is to use an in situ network of automatically-controlled pan, tilt and zoom (PTZ) cameras, and personal position and orientation sensing devices, to create static three-dimensional (3D) environment models, dynamic 3D participant models, and dynamic multi-dimensional participant pose tracks that can be viewed from any perspective. The system will include automatic semantic parsing of the dynamic models and tracks, with interactive and automatic configuration, event detection, and annotation to increase the usefulness and effectiveness of reviewed material. We also plan to devise a novel set of training and instructional approaches specially designed for this set of systems and tools.
- sponsor: Office of Naval Research (ONR)
- collaborating institutions: Sarnoff Corporation and University of North Carolina at Chapel Hill (UNC)
- partners: TECOM, PM TRASYS, TTECG
- other faculty members involved: Dr. Chris Darken, Dr. Neil Rowe, Dr. Mathias Kolsch and Delta3D team
- my role: PI for the three institutions team (FY08 - FY11: four years grant)
|
|
