Camera placement

There are some considerations to keep in mind when positioning cameras for markerless capture.

• Typically, markerless software performs best with frontal and sagittal views, so cameras should be arranged in a horseshoe configuration. Make sure to place them symmetrically relative to the capture volume.

  

• Position the cameras at approximately shoulder-level and point them so that they will see the subject’s full height. If possible, every camera should be able to capture the subject’s whole body, centered in the field of view.

  

• For most applications, the captured subject should be at least 500 pixels tall, so make sure that the cameras are an adequate distance from the recording volume.

  

• In smaller spaces, rotating cameras vertically may be helpful. It may also be advantageous to place cameras vertically when the subject will be moving in place—for example, running on a treadmill. The markerless software will then orient the video correctly based on the QTM calibration file.

  

• No matter which way the cameras are rotated, make sure that there is sufficient overlap in their fields of view.