I have a room 13 feet and 9.5 inches by 11 feet.

Is this big enough to shoot for motion capture?

Elsewhere in these forums, it is recommended that you have a minimum 10 x 10 space. Realistically, this mean you cannot walk around the room, and you may even have to be sitting. The problem is that in a tight space, it becomes difficult for your cameras to keep enough useable motion data in frame. Getting cameras with a wider FOV keeps more of the action in frame, but now you're dealing with lens distortion, which is just trading one problem for another.

FWIW, my 'regular' space at home is probably about 14 feet wide and about 20 feet long, and I'm able to move a round a little bit in this space, which is probably about 5 x 5 feet of useable acting space. Frankly, this still feels pretty tight sometimes--for example, I have to be careful not to raise my arms too high. Ideally, I want to set up this system in my garage, which is quite a bit bigger but, unfortunately, this space is too cluttered right now to use for motion capture.

If all goes as planned, I will get to do a session in a significantly larger space at work tomorrow night. I think this will give me a better idea of what the optimal space size for an iPi Studio session is.

Hope this helps.

Greenlaw

_________________
Greenlaw
Artist/Partner - Little Green Dog | Demo Reel (2017) | Demo Reel (2015) | Demo Reel (2013)


Watch a one minute excerpt on Vimeo now!

A week and a half later than originally planned, but I was finally able to do an iPi Recorder session tonight in our cafe/common area at work. I'm glad I did as much testing at home as I did because, without this previous experience, I'm sure I would have been wasting my time tonight. Anyway, the session appeared to go smoothly and I think I have some good data to work with this weekend.

The space was plenty big, and my only problem was that the floor was very shiny and it reflected the overhead lights. I turned out the room lights when recording and used two large softboxes carefully positioned to avoid bright 'hot spots.' I think it was sufficient lighting for this test. I'll post an update later.

Tip: Keep a roll of gaffer's tape handy--it makes it easy to 'tie-down' the webcams and cables without worrying about ruining expensive office fixtures. Thank goodness I work at a visual effects house because gaffers tape was easy to find here.

Greenlaw

_________________
Greenlaw
Artist/Partner - Little Green Dog | Demo Reel (2017) | Demo Reel (2015) | Demo Reel (2013)


Watch a one minute excerpt on Vimeo now!

Argh! A couple of annoying problems with my new footage:

1. Shiny floors create bad reflections. I realized this after setup and tried to minimize the problem, but the reflections are still there. I think I can deal with this in Fusion using a garbage matte and the clean plate, but it's something to be aware of in future shoots. If I use this space again, I'll need to cover the floor with mats or something.

2. Way in the background is a computer screen with a screen saver enabled. This stupid screen saver keeps confusing iPi Studio's calibration tracker (the moving Windows logo must be as bright or brighter than my Maglite.) I'll have to mask out the computer screen using Fusion and save out a new clip for iPi Studio.

3. My calibration light moves off the two side cameras at one point. It's probably okay, but we'll see.

Think I'm done for tonight; will tackle this again tomorrow evening.

Greenlaw

_________________
Greenlaw
Artist/Partner - Little Green Dog | Demo Reel (2017) | Demo Reel (2015) | Demo Reel (2013)


Watch a one minute excerpt on Vimeo now!

Does this confuse the marker tracker? We tested and tuned marker tracking in current version with shiny floors and it worked more or less good in tests, but maybe our floors were not shiny enough. This kind of problem can take lots of tweaking. I would appreciate if you could send me calibration video with shiny floor. Tuning the correlation mask we use for marker tracking may help to minimize this problem.



Sorry, our calibration software is not yet smart enough to filter out blinking computer screen or LED. We recommend turning off screens during calibration. And LEDs can be covered with a piece of paper or something similar.



This should not be a problem if it's just a few frames. Calibration should work correctly with up to 30% of misdetected marker positions.