Bringing Physics to the Surface (UIST-2008 Assigned Reading)

Bringing Physics to the Surface (UIST-2008 Assigned Reading)

By: Andrew Wilson, Shahram Izadi, Otmar Hilliges, Armando Garcia-Mendoza, David Kirk

Summary:
This study was directed in the way of exploring how the intersection of surface technologies coincides with advanced games physics engines. The surface technologies being discussed are those that are capable of sensing multiple contacts as well as some shape information. A lot of work has been done recently on interactive surfaces as well as the need to provide richer and more realistic 3D images. In order for this to be possible it was necessary for a physics engine that allowed an appropriate interaction with virtual objects. They used the following strategies to describe these interactions: Direct force (contact point of force touches a virtual object), Virtual joints and springs (contact connected to object by a link so it is dragged behind it), Proxy objects (objects near the contacted object which receive friction forces), Particles (shape information available here), and Deformable mesh (an approach to model 2D of 3D shapes). There are two strategies for moving objects. The first is by applying the external force directly and then updating the position of the object to show what is being done. The second strategy is to attach a rope to the object and move it indirectly. To further understand the utility of the techniques mentioned a study was done where participants were instructed to do 3 simple physics tasks. The researchers analyzed the various behavioral and experience aspects of interaction during these tasks. The results of this study showed that even though the more familiar approaches offered more predictable control in the study, the particle proxy approach can offer good performance as well, but with new modes of interaction (example of cupping the ball). One problem with the system is that there is no way to determine how hard, or soft the user is interacting with the objects (no way to establish a magnitude of force desired). It was also discovered that grabbing an object virtually based off of contacts on its edges is very difficult. In the future the researchers wish to improve the basic sensing techniques of the system so that things like grabbing an object are easier.

Discussion:
I thought this was a very drawn out and dull paper. I think it is an interesting area of research, but that it was really not very well organized or something was wrong with the paper. It seemed to jump around a bit. I think the idea of being able to incorporate interactive surface input into a real time physics simulation was cool. I also think that it could be modified to make advancements into a more virtual reality in the world.