I've tried some slow motion dives using 6 cameras, and dives seem feasible, but ... A full speed dive should not exceed the speed of a run by much, so 60 fps should be enough?
If so, it leaves the issue of not damaging myself ... At 6'4", 190 lbs, I land pretty hard. I see 4x8 ft, 12" thick training mats, which might work for a dive to a dead stop, like a volleyball dive in sand. Kind of pricey, though. Thick enough? overkill? And, if I want dives with some slide to them .. Think volleyball on a wood court, how thin can I go without damaging myself? Anybody have any experience in this direction? Wonder if I may need to capture an all out running dive, to a fat pad, finishing by standing to a TPose, and processing in reverse, then capturing a quick crouched slide (so as to not hit so hard on a thinner, slipperier pad, processing similarly, then glue the 2 motions together, for an all out sliding dive.
I forsee pain in my future.

Suggestions? Prior experiences?

Thanks,
Mike

Hi,

Well in my opionion, you actually don't really need to actually dive/slide, just do a stationary motions of what a dive/slide would look like if you did do one, then in your chosen 3D editing program or possibly even in iPi its self, you can key frame the slide length to any amount you want. True Bones guy has a youtube video where he shows something similar using iPi tools themselves, search TrueBones or MrBones on youtube, I don't have the actual link.

The dive could be done the same way to limit the impact, but you would probably want to analyze an actual dive to get the similar motions for realism to perform it stationary, using a narrow object or some sort of hanging harness apparatus (depending on capabilities to use), to support yourself for downward arm/leg traveling, could even use it for the slide action, but ground contact for a ground slide would be better maybe.

Hope this helps and Good Luck! Don't hurt yourself :)

The hanging harness, along with looking at actual video footage has definite possibilities. I talked to several people ... One who has actually done very similar captures, in a professional mocap studio, and he suggested going with 2" pads, which surprised me, with a cardboard layer on top when you wanted to slide. A sales lady at tumbltrak suggested 2" as well, with a 4" version if I was determined to land harder. So ... Splurged on mats, and I'll give it a try, that way first. Real dives/slides really do look different, so ... Hoping it works. Btw ... The guy who has actually done this always seemed to be getting banged up.

Mike

Ok, haha, not like regualr ppl dive and slide much, and when u do it hruts, even with a mat lol.

Yes, actually doing it would look better, I was suggesting the other ways to save you some aches.

You will need an area large enough to for what you stated and the cameras to be spread enough to capture it, you will be pushing the 20 x 20 area for PS Eyes to view in enough detail for tracking while doing run/dive/slide all as one motion capture, just depends on if the capture will turn out that way you expect.

A 20 x 20 feet area will really give you actual usable capturing area of 12 x 12 feet max.,as you need min.of 8 feet from the cameras, but it seems feasible if you have the space and can position the cameras good for your purposes, maybe some higher pointing down on mats also.

Good Luck!

I had to animate five digital stunt doubles in a movie last fall. I only had a day or so to do the animation so I turned to mocap studio.

In the sequence, each actor had to come 'flying' out of a natural water slide and in an underground lake. To do this, I balanced myself on a drummer's throne and acted out the sliding, falling and landing for each character. Then, I brought the motion data to work and applied them to the digital avatars of each actor. The result was surprisingly convincing for how little work this actually took (compared to hand key framing the motions, which was the original plan.)

The tricky bit was not falling off the stool and hurting myself during the capture session. :)

I posted a pic from the session here in this thread:

viewtopic.php?f=5&t=9155&hilit=+throne

I might post a pic from the movie later, once I'm sure it's been released.

1k0 did some nice 'free fall' captures a few years ago using the original iPi Desktop Motion Capture software. He did is by simply balancing on one foot. He also did some nice diving motions. You can see it here:

http://vimeo.com/22036677

Here's the trailer with his final animations: http://vimeo.com/22028464

Hope these examples give you some ideas for how to accomplish the stunt you want.

G.

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

Watch a one minute excerpt on Vimeo now!

Yeah, I forgot about his videos, good pull on that Greenlaw :)

And yeah, I use a padded stool too alot for things, works good and is basically hidden under you.

Great stuff.
Thanks for the input!

Mike

One more useful tip (probably obvious): if the stool is in the capture space when you run the Background Processing step, it will subtract the stool from the scene. Naturally, this help later for tracking. Otherwise, the stool will be included in the point cloud and can interfere with the results.

If you forget, you can go back and place the stool where you need it, and re-run the Background Processing step. There shouldn't be a need to re-calibrate the scene or make other changes to continue with the capture.

G.

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

Watch a one minute excerpt on Vimeo now!

Correct, when using Background Removal video compression, I don't know if it removes it if JPG compression is used during recording to shrink video file size, but it may.

With the new video editing feature, I can record it BR and then break the video up where I choose for takes, and remove un needed footage and re-save each as a smaller file in the same or different compression also, but never save lower than 75-80% JPG or tracking my suffer.

You need to edit video before you run tracking, as each will be considered a differnt video file after editing, but still uses the same calibration file.

So far, it has worked fine.

I am basically just throwing out the garbage takes I don't want or need per recording.