Using 3D Content in Futuristic Scenarios
In the future, 3D visual content will be an important
type of media that will be acquired, transferred,
and manipulated in the swarm. But are
the current approaches and tools for 3D content
processing ready to take on the challenges of the
forthcoming technological revolution?
Imagine the following scenario. You are shopping
for a teapot to give as a gift to your friend.
Knowing her preferences, you are looking for a
teapot shaped like a elephant, similar to the gurine
she brought you from India (see Figure 1). You
start by describing your gurine to the shop assistant,
explaining the details and the overall shape
it should have and using gestures to indicate its
overall size, in an effort to convey your mental
image to the shop assistant. She presents you with
a number of teapots that correspond to her interpretation
of your description. You grasp some of
the teapots to test the ergonomics of the handle.
Now, imagine yourself making the same request
to a computer avatar in an all-digital shop while
touring a virtual version of New York City’s 5th
Avenue. Would you expect to be similarly understood?
Can you imagine interacting with the digital models before deciding which one you want to
be printed and delivered to your home? Probably
not. Why?
I argue that such a virtual shopping scenario
is not possible yet because, until now, modeling
shapes has only been considered a synonym for
modeling the geometry of shapes. That is, the
meaning of objects, their function or intended
usage, and their role in a more complex scene
have been neglected, or at least only considered as
an add-on. Rethinking the whole modeling pipeline
with semantics instead of geometry as a key
driver could help develop novel representations
able to address the challenges of future cyber physical
worlds.