I haven't tried the latest build yet but I'm excited about some of the features mentioned in the change log. I hope to try the new Recorder tonight.


THANK YOU. If I understand the description, that's a very nice improvement and something I've wanted for a while now. :)


That sounds useful but I'm now wondering if it's possible to also add a grid overlay to assist in camera alignment. I often mark the spot on the floor where the actor assumes the T Pose at the start of recording, and I try to align my Kinect sensors to that spot. It's typically located in the lower third of the display, so a center target is not so useful for this purpose. If we had a grid overlay, it would be easier to figure out the alignment for markers or items that are not located in the center of the screen. Thanks in advance for considering this feature request.


Very interesting. This makes me want to try my older computers with Recorder again. In previous tests, they were just shy of being suitable slave computers. I wonder if the latest optimizations will make them more suitable.

I'll post anything new I learn from testing this build.

Thanks again guys!

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

Watch a one minute excerpt on Vimeo now!

...

Sounds interesting, I will try it out soon.

...

Alignment grid is included in Recorder v3.3.1. Hope it is close to what you asked for.

Thanks! I will check it out this evening.

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

Watch a one minute excerpt on Vimeo now!

...

The grid overlay can be helpful in set up, I think more for Kinects than PS Eyes though.

I think Greenlaw was thinking of the same type perspective grid as is used in the Studio, but I think that would be impossible on the Orthographic view used by cameras, or sensors.

The overlay added won't really show the center point of the floor, so you would still need floor markings for center point and capture volume extents, and most likely an object placed at that center point to align the cameras best in each view port, especially when using PS Eyes all set at different positions.

Nice add though, I am sure it will be used and can be turned off when not needed anyway.

...

That would be nice, but, nah, I wasn't thinking that ambitiously. I know the Kinect has ground detection but I don't think the perspective grids would necessarily line up from device to device.

I was just thinking of something like a standard 12-field grid where I can visually check if a spot or item on the floor was in the same relative x/y position for each device. Once visually 'aligned', this helps me judge if a device position/angle should be moved to improve space coverage/overlap. Sort of a visual 'pre-calibration' aid. Hope that makes sense.

Sigh! I meant to do a lot of iPi testing this weekend but got sidetracked helping my daughter with her Halloween costume. At least that's now off my plate and I should be getting back to this project tonight.

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

Watch a one minute excerpt on Vimeo now!

...

Oh ok, that's what they supplied, it should be helpful in some respects like you stated and does give a direct center line which is what I would use for the PS Eyes to all be close in each view port.

Although, all my cams are fixed in position in my capture area, but if I ever need to move them again, this will come in handy.

Thanks for the addition!

...

Oh, geez...I've been so busy, I forgot to thank the developers for this feature.

Yes! Thank you, guys. This is exactly what I was asking for. It may seem like a simple thing but it's very helpful. :)

I just updated to 3.3.2.57. Will be using it tonight if I don't come home too late from work.

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

Watch a one minute excerpt on Vimeo now!

Re: improved recording performance. I finally got around to trying Kinect One with my daughter's laptop again. Unfortunately, getting only 25 - 27 fps and plenty of dropped frames. I think this is because the laptop has an i5 processor. (Argh! I know we should have sprung for the i7! Oh, well.)

So I still haven't been able to do a successful triple Kinect One recording. Guess I better get serious about buying a new workstation--my current one is maxed out and getting creaky. (Geez...it must be over eight years old now.)

On the bright side, dual continues to work great and I'm back to finishing that Brudders short this weekend, and creating new tests for our next iPi mocap movie (finally!)

I'll start posting progress reports on our website soon, and will make the announcements here of course.

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

Watch a one minute excerpt on Vimeo now!

Another followup: Even though the performance update did not help with my daughter's i5 laptop, I'm seeing a small but notable improvement on my old workstation.

Previously, I've had to leave out the RGB data because I was getting some dropped frames but, with the latest version of Recorder, recording with the RGB video data enabled doesn't appear to be a problem.

If you're using the flat panel method for calibration, that's not a big deal, but it's good to know I can record the RGB video again. This may be useful for demo videos and tutorials, which I'll be working on soon. (Note to new users: If you're using the penlight method, which is required for three or more Kinect sensors, you will need to record RGB data.)

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

Watch a one minute excerpt on Vimeo now!