forums.naturalpoint.com

Hi there,

I am new to this really, I currently use a markerless mocap package and its time to invest in a setup like the ones offered here.

I have a couple of questions really, my first is about size of capture area (well setup really) I would like to use my living room to start with as im very small scale and this is hobby level at the moment.

My living room is about.. 12ft X 14ft so do you think I could get an 8 cam setup in here? (i understand I will not be able to have a huge capture area, but think a 6x6ft capture is realistic?

I could get more room if i used the other room that joins into the living room, but the issue is it has a partition wall sticking out.. ill try and draw the room (sorry for my rubbish drawing! haha)

http://imageshack.us/photo/my-images/338/layoutaf.png/

My 2nd question is do you people with V100R2 who then bought Flex 13, was it a wow moment? is there a huge difference? (8 cam setup maybe 6)

Thanks for your time

bit of a longshot but does anyone have raw data examples of a 6 or 8 cam setup for flex 13 and / or v100 R2 cams they would be happy to share? as the ones on the site here are 13

Hello -

I'll pass this request off to our marketing team, if there's anything available they should have it.

ok thanks Mike. Just let me know as it would be good to see

Hi Tomuk,

We don't have sample data for you to compare in those configurations, but I'm working on some images/math for you to reference which should help.

In the interim, and in short--the V100:R2's narrower FOV is going to make capturing data in that space difficult. Even the Flex 13's 56� � 44.8� FOV will yield a small volume in such a compact space, but it will certainly be larger.

Once I have the numbers and diagrams worked up, I'll post them here for your reference.

- Seth

Hi Tomuk,

There are a few basic principles that together determine approximate capture volume size. These principles are influenced by setup area (L�W�H), camera model + lens FOV camera count, and tracking application (single marker vs. full body, for example, require different degrees of camera overlap).

The estimates below assume 8 Flex 13 cameras, positioned in Portrait mode (typically not recommended, but occasionally necessary in very small spaces with low camera counts), in a 12' W � 14' L � 8' H volume (with the cameras aimed down to see 7' H on the opposite side of the volume)

Portrait mode (determined by imager, not case):



Camera/volume gutter:

The gutter is the horizontal distance from where the camera is mounted to where its FOV intersects with the ground plane. In simpler terms, it's the area between the camera and the beginning of the capture volume.

The size of the gutter is determined by the camera's FOV, the height it's mounted at, and the angle it's aimed at (typically straight across vs. downward). Depending on these factors, below are some possible gutters for your setup. (The blue line is your gutter, and the red lines are the camera FOV.)



Camera overlap:

You generally want at least three cameras viewing a marker at any time for reliable data. Below is an overlap map of an 8 camera volume in your room's dimensions. The gray areas have one or two camera overlap, while the pink areas have three ore more. (This is what the volume would look like for a single marker, or a simple cluster of a few markers--not full body data.)




Full body volume (overlap minus gutter):

The approximate full body-capable volume is determined by the camera overlap, minus the gutter. This results in a volume that has three or more camera overlap for each marker, from the ground up to 7' in the air.



In this image, the dotted lines denote the gutter (where the line ends, the camera begins seeing the whole vertical volume). The green section marks the full body volume. Here you're looking at approximately a 4' � 2.75' volume.

(In reality, you can set the software to only require two markers to solve, which will bump out the volume slightly. I would guess that you can maybe squeeze out another foot on either side... maybe.)

At the end of the day, the best thing would be to locate a new capture space. What you're currently working with is just too small. I never recommend a setup area less than 14' � 14', and that's the bare minimum. 16' � 16' and beyond would be much more ideal. An 8 camera Flex 13 volume can easily produce a 10' � 10' capture volume, but you need a larger setup space. 20' � 20' would be perfect, if you can manage it.

I hope this helps. Feel free to ping us with any further questions.

- Seth

hello,

thanks very much for your reply, so looks like my living room just wont cut it!

I would like 8 min Flex 13's but its a bit over my budget right now and am looking at with 6 flex 13's or 8 of the V100:R2 cams.

I am thinking 6 flex 13s as then I have the better cam and I would then add more as required.

Shame there is no examples of a 6 cam setup laying around, do you think a 6 cam setup is just not going to be enough to capture some movements such as 1 person sword/fighting/boxing movements and 1 person dancing for example? would 6 flex 13's give much more than say 8 R100's?

tomuk wrote:I am thinking 6 flex 13s as then I have the better cam and I would then add more as required.

I think this is a good plan. The Flex 13 is really far superior, and it will make a much better platform to expand on later on.

Shame there is no examples of a 6 cam setup laying around, do you think a 6 cam setup is just not going to be enough to capture some movements such as 1 person sword/fighting/boxing movements and 1 person dancing for example? would 6 flex 13's give much more than say 8 R100's?

The difference between 6 and 8 Flex 13s will be pretty significant, so if you can upgrade down the road, you'll be much better off for it. In the meantime, 6 Flex 13s will still be able to track one actor + prop (though nothing too acrobatic, I'm afraid)--just not in a setup area the size of your living room. To get the most out of your system, try to find a 20' x 20' space with room to mount the cameras about 8' in the air.

For more complex movements, or for two actors, you'll be looking at something closer to 12 - 16 Flex 13s.

OK thanks again, no not in my living room haha, im going to have to find a new place. I was just trying to save money by using what I had but clearly that is going to really limit me.

SOrry the 2nd question was if you think 6x Flex 13's would be better than 8 R100:V2 or even 10..

Reason I am asking is I have seen that I could even get 12 R100:v2 for the price of about.. 12 R100:s so, what im really getting at is i know if I maxed out the rigs on 24 (is it 24?) cams then having 24 Flex 13's is better than 24 R100's, but when your looking more mid numbers considering spend and what you get for it, whats the best bang for buck?

6 Flex 13's or 10/12 R100's?

THanks

If you were just going to purchase a system and stick with it, with no upgrades later on, I'd probably recommend a 12 camera V100:R2 system with 3.5mm lenses. That'll likely give you the most volume with acceptable data at that price point.

But if you want to look toward future upgrades, the Flex 13 will start to seriously outshine the R2 from 12+ cameras on. For most applications, I'd take a 12 camera Flex 13 system over even 24 V100:R2s.

If you can afford the $10.5k for a 12 R2 system, you might consider saving up for a bit longer until you can reach the 8 Flex 13 system at $11k. That'll be a great system, with tremendous potential for expansion down the road.

But if that's just too far out of reach, yeah, go for 12 R2s. For your application, the increased camera count will probably make up for the loss in resolution.

- Seth