Page 1 of 1

GDC virtual camera demo

Posted: Wed Mar 17, 2010 10:40 pm
by Jim
Hey all:

Here is a link to some coverage our new Virtual Camera system got when we showed it at GDC this past week.

http://www.engadget.com/2010/03/11/opti ... r-delicio/

We are continuing work on the project and hope to release it around Siggraph time.

Looking forward to your comments and suggestions!

Re: GDC virtual camera demo

Posted: Fri Mar 19, 2010 4:46 pm
by Davve
Great!

looking forward to this.

There is only 2 more things on my wishlist now.

That single cam attached to the head for face cap, avatar style.

In the vid he says that he is in a differnt cap volume. why can�t he walk from one capture volume to another? That would be a nice feature.

Keep up the good work!!

David

http://davidaberg.se

Re: GDC virtual camera demo

Posted: Fri Mar 19, 2010 4:59 pm
by Seth Steiling
Hi David,

Trade shows impose constraints on our volumes that you wouldn't necessarily be subject to at your own facility. (Things like booth size, attendee traffic flow, space for video displays, etc.) The virtual camera and mocap actor actually can both exist in the same volume. We just split it up at the show. Of course, more cameras will give you more accuracy and a larger volume, both of which are ideal for doing simultaneous mocap and virtual camera work. So, you might want 18+ to do both at the same time.

Re: GDC virtual camera demo

Posted: Wed Apr 07, 2010 11:44 am
by JustinDenton
Hi Davve,

I was the guy doing the demo in that video. At our offices, we have and do capture both character and camera in the same volume. It's so much fun to get up and physically do a camera move instead of keyframing it!

As for the walking from one capture volume to another, I think that Tracking Tools is currently their only software that can handle this, and if you are doing a character capture, you will be using Arena instead. Maybe some of the OptiTrack guys can correct me if I am wrong on that issue.

Also, as Seth stated, we are using an 18 camera setup to achieve the results we want to minimize occlusion from the camera operator to the skeletal solve.

Cheers,
Justin Denton