Support

Motion Capture FAQs

1. General

1. What are recommended accessories for motion capture?

• OptiTrack Motion Capture Suit
• Motion Capture Marker Sets
• Camera Mounts

2. What type of motions can be captured?

The ARENA software line has the fidelity to capture all types of human body movement including both subtle and fast motion. Facial movement and skeletons can be tracked as long as a sufficient number of markers are visible; some positions may require adjustment to the marker set and camera placement. More dynamic motions will benefit from a larger capture volume to work within.

3. Is sample data captured from the system available for review?

Sample data captures are available in BVH and C3D formats from the video section.

4. Is it possible to use props and have them captured along with the skeleton data?

Yes, props with 3+ marker configurations can be captured as rigid bodies.

2. Tracking

1. Is ARENA software capable of real-time tracking and skeleton solving?

Yes, real-time preview with skeleton solving is supported. This includes the ability to stream live skeleton data to MotionBuilder using a plug-in.

2. How accurate is the tracking?

The 3D location of markers can be resolved with millimeter accuracy and resolution depending on capture volume size and camera configuration. Increasing the number of cameras can help improve the tracking performance if needed.

3. How many actors/skeletons can be tracked at a time?

Two actors/skeletons can be tracked at the same time. Support for more than two actors is under development.

4. How does ARENA tell the difference between multiple actors/skeletons?

ARENA uses the position of the back waist marker to identify different skeletons. For two actors, we recommended that one skeleton have the waist marker offset to the left of the spine and the other have the waist marker offset to the right.

5. Are custom skeletons supported?

ARENA does not support importing skeletons, but it is on the list for future versions. For animation in Max, you could apply BVH data to the model, or you could go through MotionBuilder and rig the C3D data from ARENA to your skeleton.

6. Is capture of facial movement and fingers supported?

Capture of facial movements is possible with the ARENA Expression software. Support for individual finger tracking is still under development.

7. What capture volume sizes are supported?

Camera setup areas of up to 20' x 20' (400 square feet) are supported; this typically produces a usable capture volume of 12' x 12' (144 square feet).

8. Can fast motions like baseball bat and golf club swings be captured?

Yes, our mocap cameras are capable of shutter rates up to 1000 and higher Hz, so it is possible to capture non-smeared motion for most objects. A golf club head travels at about 125 MPH, and that can be captured. It would have a large gap of about 18" between samples, but the motion is linear at the fastest point, and could be interpolated if desired.

9. Can facial mocap be captured at the same time as full body mocap?

Camera configurations which are ideal for full body mocap are usually not suitable for face mocap, so capturing face and body motion at the same time is not currently supported. Support for capturing both at the same time is under development.

3. Software Formats, Exports, and Streaming

1. What export formats are supported for motion data?

Body data from ARENA can be exported to BVH, C3D, and FBX formats and streamed via real-time plug-ins to MotionBuilder, 3ds Max, and DAZ Studio.

Face data from ARENA Expression can be exported to C3D and FBX formats and streamed via real-time plug-ins to MotionBuilder and DAZ Studio.

2. What is the frame rate of the real-time streaming from ARENA?

The real-time streaming rate to MotionBuilder (or other NatNet clients) is based on the skeleton tracking rate, which is often derived from the native camera frame rate (100, 120, 250, etc. FPS).

To achieve the best performance, ARENA should be run on a different computer than the application (MotionBuilder, etc.) to which the data is being streamed.

3. Can the real-time streaming skeleton data be recorded using the MotionBuilder plugin?

Yes. View the MotionBuilder Plugin User Guide for details.

4. Can NatNet stream real-time motion capture data over the Internet?

NatNet does not support streaming over the Internet. A direct ethernet cable connection or private LAN/hub for the client and server are the preferred configurations. Streaming over local LANs is also supported by NatNet, however that setup may generate unwanted traffic for other users on the network.

4. Hardware and Setup

1. Which cameras can be used with ARENA software?

See the software comparison table for more information on which cameras are supported by each OptiTrack software application.

2. How many cameras do you recommend, and what is the maximum number?

Six cameras in the minimum for single actor full body motion capture, and eight are recommended for improved tracking and a larger capture volume. For two actors at the same time, 12 cameras is the minimum. The current limit is 24 cameras. Support for more cameras is under development.

3. What is the best way to mount the cameras?

Cameras should be setup so that they do not shake and are not easily moved. If any camera is moved once calibrated, tracking accuracy may be affected and the cameras should be re-calibrated.

Suitable mounting hardware includes tripods on concrete floors, truss work¹, and wall mount brackets. Also, take a look at the volume layouts located on our bundles pages—and elsewhere—to get an idea on where to place your cameras.

4. What are the system requirements for motion capture?

See the minimum system requirements listed on the ARENA motion capture software page.

The Foundation Package and recommended accessories provide you with the building blocks for your motion capture studio.

5. Can the motion capture data be synchronized to external sources?

When using and OptiHub or eSync, cameras can synchronize their exposure timing to external sources for use as single frame or free-running triggers. Common uses are to synchronize capture data with software controls or hardware such as force plates and shutter glasses.

6. What is the maximum distance a camera can be used at?

The maximum cable length for a direct USB connection is 5 meters (16 feet). Hubs and active extension cables can be used to increase the range of use. The recommended configuration for large volumes is to connect hub(s) to the computer using active extension cables, then connect cameras to the hub(s).

5. Scene Audio/Video

1. How is scene A/V set up and recorded?

Camera Setup - To select a camera, drag the camera icon from the Cameras node of the Asset window into the Scene A/V Cameras node. To select a DirectShow video capture source, select 'Add' from the Scene A/V context menu (right-click on the Scene A/V group node in the Asset Manager).

Viewing - To view scene A/V click the view name text in the upper left-hand corner of a 3D view and select the camera name from the drop-down list.

Recording - Scene A/V data is automatically recorded when a take is recorded. Recording can be toggled on/off using the Scene A/V Camera context menu (right-click the Scene A/V camera item in the Asset Manager).

2. What video and audio sources can be used for scene A/V?

OptiTrack cameras can be used to capture scene video.

3. What are the benefits of using a calibrated camera for scene A/V?

Using a calibrated camera allows reconstructed 3D markers to be correctly reprojected onto scene video.

4. How does scene A/V sync video data with motion capture data?

OptiTrack cameras will capture scene video synced with the Motion Capture data on a per-frame basis.

Files recorded from generic DirectShow devices, such as web cams, DV camcorders, etc have variable recording rates. Playback will use the framerate of the file to determine mocap to video frame matching, however because the recording start time will be different, the start offset must be synced manually upon playback using the scene A/V File's 'Frame Offset' property (accessed from the scene A/V File context menu).

5. What format does scene A/V record to?

Scene A/V records to motion jpeg with an AVI wrapper which includes audio. Audio is recorded uncompressed at the sampling rate of the source device (configured in the OS and/or DirectShow source filter's property page).

6. What configuration provides the best bandwidth for scene A/V?

Make sure USB video sources are on their own USB EHCI controller, this will isolate their bandwidth consumption from cameras capturing mocap data. You review this by checking the camera connection tree in the Arena/Expression Asset window, or the Windows Device manager using view by connection).

7. What if video is not working for a OptiTrack V100 camera?

First, try to make sure the OptiTrack camera is on its on USB EHCI controller.

Then, if video is still not displaying, and you suspect USB bandwidth to be the cause, you can adjust the camera's scene A/V properties as follows:

Asset Manager -> Select Camera -> Right-Click -> Properties

Framerate: 25 FPS
Size Decimation: Quarter Size
Frame Decimation: Send Every Fourth (or lower)

1. NaturalPoint recommends truss systems from Milos Structural Systems and Trussworks.