If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Hi Everyone. I am trying to find out what size of lab space people have for motion capture. Could you help my by posting the dimensions of your lab, and if known what size of functional capture area you get?
Luckily enough - huge ;-).
Actually 25 x 6.5 x 7 m, with 10 cameras mounted on a gantry 2.5 m above the floor it is possible to get approximately over 50 cubic meters of calibrated space, ample for most sporting actions.
Bruce MacWilliams, Ph.D.
Associate Professor, University of Utah Orthopaedics
Director, Movement Analysis
Shriners Hospitals for ChildrenĀ®
Salt Lake City, UT 84103
801.536.3800
Functional capture area will not only depend on the physical dimensions of your lab but also the type of cameras you are using. Camera variables include:
Number of cameras - ensures less chance of marker drop outs, but obviously allows you to expand capture area provided you have physical space in your lab to do so.
Capture frequency - the camera field of view (FOV) will shrink as frequency increases, so depending on what you want to capture will affect your capture area, if you are capturing something slow moving at a low frequency you will have a much larger capture area. If you are capturing something fast at a high frequency you will have a much smaller capture area.
Focal length of the lens on the camera - Some camera suppliers (ex. Vicon) allow you to pick and choose your lens which allows you to optimize your capture area a bit depending on the shape of your lab space, other camera suppliers (ex. Qualisys) come with a set lens focal length which you can't adjust. A 12.5 mm lens will give you a wide FOV but the camera will have to be closer to the subject. A 25 mm lens will give a narrow FOV, but allow you to see markers from much further away. If you had a long rectangle shaped room, you would probably want cameras with 12.5 mm lenses on both sides and 25mm lenses at both end.
Diameter of the markers you use - small markers usually require the cameras to be placed closer which give you a smaller capture area, larger markers give you a larger capture area. But as I explained above the focal length of the lens can help with this, ex. you can see a small marker from far away.
Brian Schulz, Ph.D.
Scientific Program Manager for Rehabilitation Engineering and Prosthetics
Rehabilitation Research and Development Service
Office of Research and Development
U.S. Department of Veterans Affairs
810 Vermont Avenue (10P9R)
Washington, DC 20240
Email: Brian.Schulz@va.gov
Phone: (202) 443-5769
My lab is 37' (11.3m) long x 31' (9.4m) wide x 9' (2.7m) high and our 13 camera system can generally resolve markers within a central 16' (4.9m) x ~6' (1.8m) area. Marker visibility rapidly deteriorates near the edges of that central area though. I'm also not too sure about the width of my capture area, as people generally stay within a 4'-wide walkway.
Chris's comment about camera lenses is well taken- my cameras on the longer side walls needed wide angle lenses while the cameras in the corners and on the shorter end walls needed a tighter focus; however, his comment about capture frequency is not true about all camera models. Some do indeed "window" as he describes at higher capture rates while others do not.
Hi Everyone. I am trying to find out what size of lab space people have for motion capture. Could you help my by posting the dimensions of your lab, and if known what size of functional capture area you get?
Thanks!
Hi Chris,
my lab is 11.4 x 7.2 m, but I have a column and a non usable corner. The 10 cameras are not regularly spread and I use 12.5 mm lenses for all cameras but one on the long end which has a 25 mm lens. Camera height varies between 2.20 and 2.60 m. Six cameras are 1.3 MPx and 4 more cameras are 2 MPx. The volume I cover is roughly:
5 x 4 x 2 m (XYZ)
If you are doing a survey of lab sizes then this response will not help. However you have asked an excellent question pertaining to lab size and functionality. jdeterman16's response is a good response and I'd like to add to it.
As a sales person that has sold and set up 100's of mocap labs, the final answer is there is no standard room size. I have seen labs in ice rinks and skating ovals, gymnasiums, mocap studios, olympic training centers and now outdoor with no walls to rooms 10x 12x 8 ft. for facila and hand capture -32 x 20 x 9 for gait, 32 x 20 x 12 for golf, 100 x 50 x 24 for baseball.
The main question is: what are your applications? Once identified you can then determine your spatial needs to capture the desired outcomes. To see a baseball pitch's result you need a room 75' long to determine if its a strike or ball ... but is that critical to determine the 8 events of baseball throwing? Can the pitcher throw into a net 20 feet away? Can a golfer be assessed in a room 32 x 20 x 12 and hit into into a screen or virtual setting to determine ball flight while their actual activity is captured in a 10 ft cube?
Or in gait if you only have a 15 x 10 x 8 foot room what do you do? Treadmill, Single sided or fight with your 3D system to do bilateral?
So a rule of thumb would be to get full height of a subject: will need a distance to camera of at least 7 feet from your desired field of view. Your camera variables are: Focal length, Lens type 12mm (very wide) to 50 mm narrow which give you a horizontal and vertical 2D FOV, and ability of your camera's light to see the desired size of marker which is broken down into pixels and subpixels.
If your room is 30 x 20 x 9 your field of view would then be 16 x 6 x 8 (Optimization of the camera's sensors - pitching / rolling / tilting the camera to map the volume can increase the volume) - that would give you 4 + gait strides, single person. and probably need 8 cameras for very good results and minimal editing.
If you can afford only 6 cameras then test a marker set and you may need to adapt a different marker set design to accommodate your application.
If you have four cameras consider single sided capture versus bilateral. Great data with single side capture vs heavy gap filled with bilateral and four cameras - but you decide.
If you have two cameras - keep the volume small and at a focus task.
if you have the luxury of space, financial resources and require multiple objects to be tracked - then 12, 16, 20+ cameras will be required to create multiple zones and to minimize occlusion.
Comment