July 20, 2011

SynthNav is the second LVC training project continuing funding from the Veldt project consisting of three parts: Open source after action review, 360 degree camera system for trainees, and Multivew3D.


Brice is the project lead and lead developer for this sub-project of the SynthNav project. The goal of Multiview3D is to design a system where each participant within the Veldt environment, typically two to four people – possibly more – sees a display in 3D or stereo from their correct viewpoint. For large stereo displays such as movie theatres, auditoriums and some immersive projector systems a static viewpoint is assigned to the display. While this is often adequate for large audience viewing, when trainees see this static screen they are at a decreased tactile advantage because they do not see their “true” view into the virtual world. See Figure 1 for a visual description of the viewpoint situations:

Figure 1: Viewpoint Differences

As you can see, without displaying a persons accurate view of the virtual objects it is possible for them not to see objects that can see them in the virtual world. If that object was an opposing force entity, it would provide unnatural tactical advantage to the opposing force entity. While head tracking works well for one user for displaying their accurate viewpoint,  head tracking of multiple people is more complex. Multiple experiments have been conducted investigating distortion present for non-tracked viewers when a static viewpoint is defined in a stereo CAVE environment, one such study is shown in Image 1.

Image 1: Non-head Tracking Distortion Study


M. Burton, B. Pollock, J. Kelly, S. Gilbert, E. Winer, and J. de la Cruz, “Diagnosing perceptual distortion present in group stereoscope viewing,” Proc. of Human Vision and Electronic Imaging XVII, IS&T/SPIE Electronic Imaging, 2012.