The EDITORIAL Workflow - 3D Postproduction: Stereoscopic Workflows and Techniques - page 69

3.2.3 Conform
As with 2D conform, 3D conform consists of swapping low-resolution (or
proxy) imagery for high-resolution (or source) imagery. Once the imagery
has been edited in proxy format, the team links to high-resolution versions
of the imagery for creating i nal deliverables. 3D conform can be part of a
“digital intermediate” (frequently abbreviated as “DI”) or “i nishing” process
along with i nal geometry, depth and color work (Figure 3.2) .
Whether or not your show has a conform step will depend on the resources
available and the kind of show you shot. In some cases you can skip the
conform process entirely and accomplish everything required with editing
only. For example, you can shoot 3D imagery with a GoPro 3D setup,
cut it together in Adobe Premiere using the Cineform codec, upload it
to YouTube using the YT3D l ag and have it in front of viewers within a
matter of hours or even minutes. For some 3D shows this may be perfectly
acceptable.
Some editing systems, for example Adobe Premiere and Avid Media
Composer, can work with high-resolution source i les as if they were proxies.
This solution avoids the intermediate step of creating (and tracking and
storing) proxy imagery and metadata and the possible risks that come
from this approach, such as stripping out source timecode and losing the
ability to link back to hi-res and/or source during conform. However, these
systems also require additional hardware such as graphics accelerators to
make real-life, real-time playback of high-resolution footage possible. If
you have the budget and need the fast turnaround these systems provide
then you should consider them for your show. However, in this topic we
assume that your show requires a conform step for at least one of your
show deliverables.
3.2.4 Compositing and Multiple Video Layers
Some teams may add additional video layers or graphics to live action
imagery as part of i nishing. These can be extremely effective as evidenced
with Catherine Owen's remarkable work in moving between layers of 3D
imagery on the 3D concert feature U23D . So although we speak about video
in this topic as if there is only one layer, remember that your show can
contain multiple video layers, all of which need to play nicely together in
Z-Space (i.e., generate neither discomfort nor discontinuity). We will see
in Chapter 7 that layers containing computer-generated elements as visual
effects also need to coexist in Z-space without creating discomfort and/
or discontinuity.