|
|
|
| e-Pub |
Previous | 
Home
Section: New Results
One Reality
Participants: Joan Sol Roo and Martin Hachet
This project explores the combination of Physical and Virtual Reality through the usage of Mixed Reality. Early explorations involved the usage of Spatial Augmented Reality in combination with Virtual Reality, two technologies with complementary characteristics that evolved separately in the past. Spatial Augmented Reality (SAR) augments the environment using projectors or screens, without the need of user instrumentation. By keeping a single unified frame of reference, it supports social interaction and natural perception of the space, but the augmentation is limited by physical constraints (e.g., it requires a surface to display information). Immersive head mounted displays on the other hand are not limited by the physical properties of the environment, yet they isolate the user from their environment. We have proposed a unified frame of reference for both SAR and immersive displays, where the users can select the visualization that is best suited for a given task (Figure 8). This enables both asymmetric collaboration between users, and back-and-forths for a single user. These explorations were followed by the combination of additional modalities, in an incremental fashion. This way, one or more users can chose the desired modalities, and immerse themselves as much as the task requires. As a result, the virtual world can be framed in relationship with the physical one.
A preliminary version of this work was presented at 3DUI 2017 [30] where we obtained a best paper -honorable mention- award. An extended version of this work was then presented at UIST 2017 [33].
In order for such systems to succeed, it is required that users are able to create unified mental models out of heterogeneous representations. To better understand how humans perceive hybrid systems as the one described above, we conducted two studies. They focused on the users' performance on heterogeneous systems (using Spatial Augmented Reality and immersive Virtual Reality displays), and combining viewpoints (egocentric and exocentric). The results show robust estimation capabilities across conditions. This work has been (conditionally) accepted at CHI 18.
