Import Blender Animation

ChipGallo wrote on 10/2/2009, 10:55 AM
I have a 5 minute PSA which features a talking head cartoon character. I would like to have photos and other image files on one layer and a Blender animation of the character on another. The head would move around, point at the graphic and so on. Any gotchas?

I have Vegas 8, Ultimate S Pro, etc. The final output would be to Flash or possibly WMV. I'd like to also put it on DVD or maybe BD later.

No experience (yet) with Blender but please don't tell me I need a Mac to do this right :-)

Comments

FrigidNDEditing wrote on 10/2/2009, 11:42 PM
make sure you set RGBA as the output and use an alpha supported format ( .png's are my preference ). import those .png's as a file sequence, and then if you need to re-render, you only have to re-render frames that are altered from the changes, and when you go back to vegas. It's what I do, though I don't do any rigging and animating, I just do text and graphic design modeling in there. (I'm too cheap to buy Lightwave from NewTek or C4D/3Ds/Maya - which are all owned by autodesk)

Dave
JohnnyRoy wrote on 10/3/2009, 11:46 AM
I agree with Dave on using PNG's. An alternate option that I use is to render useing the Quicktime Animation codec. It supports 32bit alpha channel output and is designed to hold up well with computer animation (as the name implies)

~jr
Laurence wrote on 10/3/2009, 12:40 PM
There's a Quicktime .png format that I use all the time for this sort of thing because it is lossless, small on animation type stuff (because the .png format reduces this sort of file to very small sizes) and because it supports an alpha layer. This gives you one file instead of hundreds of individual numbered frames.

Peter: Does the Quicktime animation codec support an alpha layer?
Grazie wrote on 10/3/2009, 12:46 PM
Peter? Who's Peter? - And yes it does.

I use Lagarith with RGBA - "A" is the alpha part. I use this with great effect for Particle Illusion work.

Grazie
MTuggy wrote on 10/3/2009, 8:26 PM
I do this kind of work in Lightwave but always export to sequential PNG's - none of the video out puts match the quality of the PNG's. Good luck learning 3D modeling - its a steep curve. Have you considered just using Crazy-Talk. It would be a lot easier to get a talking head in that program than doing it with a 3D animator.

Mike
ChipGallo wrote on 10/4/2009, 6:38 AM
Thanks for the suggestions on output formats. As far as learning a 3D modeling program, I am hoping that the time invested in learning it will allow me to add some value on future projects. Our main character, Mrs. Greenmeister, would look better with animated stick arms!

I will check out Crazy-Talk in case this thing needs to be finished sooner than my spare time allows.
LReavis wrote on 10/4/2009, 10:42 PM
MTuggy - have you compared .PNG sequences with the .PNG flavor of Quicktime? I have been disappointed with the introduction of black fringes around chromakeyed images when rendered with an alpha layer in Quicktime that, in contrast, looked fine when rendered directly from Vegas. The Quicktime .PNG option also messes with the image in the same way when using CineGobs Keyer.

In Bluff Titler I can choose Logarith RGBA and get clean alpha channels, but in Vegas there seems to be no way to get the "A" in RGBA to yield a valid alpha channel.

I had wanted to render the chromakey vids with alpha channel so that I could have more freedom to manipulate pan/crop when I put those clips back on the timeline of the final project, and to avoid the heavy rendering load required by chromakeying when rendering the final project . . . and still hoping I can find a way without introducing degradation of the image. Any suggestions would be appreciated.
TheHappyFriar wrote on 10/5/2009, 6:32 AM
i've never had an issue with using sequential png's or the QT version, both have been identical in everything I've tested them with (except thousands of files vs 1).

All the guys on blender forums recommend sequential png's (or TGA's if you need that, both are identical) for one specific reason: if things crash while rendering, png's done go away, single video files normally do.
FrigidNDEditing wrote on 10/5/2009, 2:55 PM
I always use sequences in a separate folder so that the project just has a folder instead of a clip, and then if I have to change something at frame 721 that affects 26 frames in total, I only have to render 26 frames in total starting at frame 721 rather than the entire animation.

Almost always, the client will end up wanting to change something, or I'll get it all assembled and edited in Vegas and something will stand out to me that I wish was different, and rather than waiting for all that rendering, I just pump out a few frames. I do the same thing with AE into Vegas for the same reason, and it has saved me literally hours over the course of just this year alone. So while I will agree that a PNG Alpha MOV file is a bit easier, I MUCH prefer using the actual PNG files, because then I have more options.

Dave