Graphics Reference
In-Depth Information
rest3d
Content
Management
Modeling Tool
Max, Maya, etc.
Figure 2.8.
Managing content in the cloud.
There are also disadvantages such as trusting the cloud provider to keep our
data secure and initially moving our data to the cloud. But those are manageable;
for instance, data can be encrypted.
Moving workflows into the cloud requires new technologies and trade-offs.
Not having direct access to data on the local disk, asynchronous processing, and
latency are issues that need to be addressed. Before being practical and effective,
a majority of the ecosystem has to be available and interoperable in the cloud.
Cloud service providers have been very successful at providing an interface
to their data and services through an API designed to follow the REST protocol
[Fielding 00]. rest3d's goal is to research how such a cloud API can be designed
to enable a 3D content ecosystem to be moved to the cloud [AMD 14].
The first step is to enable tools to dialog with the cloud as shown in Figure 2.8.
COLLADA works well for transmitting data between various modeling tools and
the cloud. Existing content-management systems can handle XML documents, as
well as provide queries and search, i.e.,
XQUERY
, inside the document. Additional
metadata can be attached to the content and provide specific services, for example
geo-specific information added to COLLADA for 3D content placed on Google
Earth. Modeling tools as a cloud service are already available, such as http://
clara.io/, https://tinkercad.com/, and http://home.lagoa.com/.
Once the content is available in the cloud, we can move the content pipeline,
and more specifically, glTF conversion, into the cloud. Content can be processed
on-demand for a particular device's needs. This solves the fragmentation of de-
vices issue that is inherent in cloud computing and the explosion of devices that
consumers expect to be able to use to access the same content. This is mostly
solved for video and audio, where the size and quality of content streamed is
based on the capacity of the target device and the available bandwidth. rest3d
and glTF work together to enable this for 3D content. (See Figure 2.9.)
Request with device
parameters
rest3d
Content
Processing
Runtime
Engine
COLLADA2GLTF
Device-specific response
Figure 2.9.
Moving the content pipeline into the cloud.
