The InfoComm show has kicked off again this year, and I’m already excited about some of the trends I’m noticing. The show was preceded by the “Projection Summit”, a smaller gathering of users, technologists, and integrators who are focused on the projection side of displays. It seems, auto-stereoscopic displays are as hot now in the commercial world as they were five years ago in the academic research community.
The Projection Summit dedicated a session to autostereoscopic display and I was happy to see that clusters of automatically calibrated projectors may be the right way to go. The goal of an autostereoscopic display is to give the user a perception of 3D without wearing glasses. Traditonal approaches that rely on rendering two distinct eyepoints (i.e. through time or wavelength multiplexing) must be abandoned. In their stead, a whole set of interesting scientific and engineering challenges arise.
In 2006 and 2008 colleagues and I published results on some of the first projector-cluster driven autostereoscopic displays that approached the problem in its most basic form. We asked the question, can a cluster of projectors fill a volume of space with controlled light rays in such a way that you can replicate a physical light field? The answer was yes, but, at the time, cost and complexity of the engineering involved precluded real deployment. It is exciting to see that we are now very close to workable products based on similar principles.