Hello

Excuse the newbie question I did have a search but found nothing.
I am doing a shoot next week where I have an actor on set doing manual work and I need to create an effect where the shot becomes an x -ray of the actor.
I have created the skeleton in 3d and am able to rig a bvh to it no problem. This is where I came across iPi as I needed a markerless caption device because I need to be able to use the live action footage to composite with skeleton.
So I have a couple of questions.
Does this system work if the cameras are moving or do they need to be locked off?
Secondly is there any way if I attach a kinect camera beneath my canon 5d is there any way to calibrate the the two cameras so I can use live action but create my bvh and they will match. There is a system called RGB+D that I could use.
Any pointers or tutorials of how best to do this be great fully received
Thanks
H

Hello

Is there an easy way to calibrate a depth camera and a dslr to use live action footage with the bvh file.

I have modelled a skeleton in 3d and can fit a bvh to it happily. But I would like to now match the movement of the skeleton with live action and composite the two together to give the effect of a partial x-ray.

Interest is a software called RGB+D that allows you to do it just wondered if anyone had done this using iPi

Any thoughts, tips or tutorials appreciated. Is two depth camera enough to get a good track. There won't be much movement in the action. Does the camera need to be locked off?

Thanks

H

The cameras should stay in place while shooting.

Currently our software does not support the calibration of mixed configurations (depth and RGB cameras).

Hi,

This is just an opinion but I really think you need reschedule your mocap shoot.

While iPi Mocap Studio is relatively easy to learn and use, using it correctly and effectively does a bit of skill and experience. If you can't find an experienced iPi user to advise and run the system for you on set, I would recommend taking a few weeks to learn the system's workflow, capabilities and limitations before diving in during an actual production. Otherwise, in your completely inexperienced with the software and gear on shooting day, you may find yourself wasting not only your time but the time of others involved with the shoot.

As for what's needed for the shoot, it depends on the action your attempting to capture. Two depth sensors is ideal for many motions so long as the performance space doesn't need to be too large--this is what we use for most of the actions in our short films. Currently, you can only use up to two so it's not an option to add more.

Alternatively, you can use multiple RGB cameras and what's preferred is the PS3 Eye. It's possible to use a higher resolution camera but doing so will complicate your workflow and IMO, there really isn't a benefit in doing so. If you use six or more PS3 Eye cameras, you can capture motions more accurately than with dual depth sensors. Keep in mind that using the PS3 Eye setup introduces several environmental, lighting, and clothing considerations that are not necessary when shooting with depth sensors.

Again, know exactly which system will best for your situation comes with practice and experience.

Hope this info is helpful. Good luck and I hope things go will. :)

G.

_________________
Greenlaw
Artist/Partner - Little Green Dog | My Demo Reels (2013,) (2015,) (2017,) and (2019)

Watch a one minute excerpt on Vimeo now!

Thanks for your reply shoot was in March so done and dusted now . Like many things we learnt on the job. Worked out pretty good. Dewalt promo on Vimeo.com/Harico

Cool! Glad it went well. Sorry, I hadn't realized I was responding to an old post. :)

_________________
Greenlaw
Artist/Partner - Little Green Dog | My Demo Reels (2013,) (2015,) (2017,) and (2019)

Watch a one minute excerpt on Vimeo now!