No announcement yet.

Data Conditioning: Before or After DLT?

This topic is closed.
  • Filter
  • Time
  • Show
Clear All
new posts

  • Data Conditioning: Before or After DLT?

    Dear Biomch-Lers,

    I am in the process of manually digitising some movement sequences in Vicon Motus. Once I have obtained image coordinates from my two camera views, I need to take these data files elsewhere to 3-D reconstruct the movement sequence using DLT (I need to do this because of the way the video footage was taken - a multi-phase calibration procedure was used and no physical synchronisation of the camera shutters was present - which Vicon Motus cannot cater for). As Vicon Motus contains a good range of data conditioning options (e.g., Butterworth filter, cubic spline, FFT, etc.) that can be set to 'optimally' condition each individual time series (using Jackson's knee method), I was wondering whether it was efficacious to condition image coordinates before DLT? Although this is unconventional, it seems like a reasonable strategy on the face of it because the image coordinate trajectories will more likely resemble the actual trajectories of each of the landmarks, and the global RMSD over the four planes for all digitised landmarks will probably be reduced, which aids data synchornisation (after the method of Yeadon & King, 1999). It might be possible to import the object-space coordinates back into Vicon Motus but it would save time and effort if it proved to be possible to condition image coordinates prior to DLT.

    I'd be interested to hear your views.

    Best wishes,

    Paul Glazier

    Biomechanics Laboratory