Compositing Video With Cg Elements

[user_deleted]

i want to make an animation that will consist of a dj spinning turntables. The dj is going to be played by an actual person that i will rotoscope to get that cartoon look. I want to composite that footage with the some cg turntables that i modeled. I have a really general question, I just want to know what would be a good workflow for accomplishing this. I know this isnt the shake forum but i thought id mention that i have access to shake as well as maya if needed. I am not really profficient in shake though i can key out a green screen and that is about it. I was alose wondering if there is a way to import video footage in maya to line up all the elements. So bassically i know how to build the individual parts but im not sure how to put them together. Any help would be appreciated.

[fortner_mike]

Well i work with after fx, but still work flow should be similar. I'd first film the dj on the green screen, then aply any filters to make it look cartoonish and edit out the green color. Then take the 3d model and render it. Then put both into post production software, and using z-depth i would put one in front of the another. Maybe some scaling and moving to get the placement right.

[user_deleted]

..or process your real footage, export it as a TGA (or iff) image sequence and place it on your Maya scene as image Plane.
This way you can place your 3D object in the correct perspective and make sure your camera angle match the real footage. You can also play with "fake" reflections of the actors on the object.

PS. one tip : during shooting, after locking the camera, I usually shoot few frames of a cube box of known dimensions (i.e 40x40x40cm), so its easy to recreate the camera focal, angle and placement in Maya, and to estimate the relative measurements of the 3D objects.

Guyb.

[joojaa]

If you have Maya unimitted you can track those known cubes with live so n matter how the camera moves maya will be able to pick the movement up and camera moves with it.

NOW incidentally if you have live and a locked camera you could track while locking the camera for a VERY perfect fit (live fits the camera much more accurately than a human would). As a bonus you could now after locking track features on the HUMAN. so you could track the exact position of ears for example (with a bit of work grantted)

Tip just to add add a bit to the above. even if you bring in what you see into maya its only a reference, be sure to assemble it in the com positioning app and render without the bg because you get a much more flexible work flow that way. Its faster and affords you to miss a bit.

[user_deleted]

Thanks to everyone who replied. Ok after reading all you guys had to say and doing some research, this is how i think i would be most productive at this? Film my dj on a green screen do the cartoon filter stuff and then edit out the green screen in shake. now this is the part that i am not sure is possible. Export the camera data from shake to maya live for the purpse of matching up the camera moves. That workflow seems to make sense to me if it is possible?

[user_deleted]

ok i had it backwards. I figured out that there is a multiplane node in shake that brings in maya camera track data that i got with the 15 day moujou trial. Now too the shake forum!