Wavefront calibration over a field of view

[tpr0p]

Measure/estimate the optical transfer function between the SLM and the camera. Has been explored by Kavakli et al..

[ichristen]

There’s a bit to talk about here.

• First, odak operates in a different regime than we do. Look at Fig. 1 in the linked paper. Their hologram is actually a nearfield hologram filtered by a pinhole. This problem is a much easier problem than the farfield holography that slmsuite focuses on (though it would be cool to add nearfield stuff too). The transfer function is much simpler because it is nearfield.
• The current wavefront calibration routines in slmsuite calculate the farfield kernel (corresponding to the phase correction / measured amplitude plus a FFT) at a single k-vector. This is valid at this single k-vector, but the full complex kernel requires the phase corrections for every k-vector.
• We discussed, in a separate channel, the slmsuite roadmap to resolve this. Maybe coming next release!
• At one point, I thought that taking odak's approach was the way to go (Add different transfer functions #32), but since changed my mind to take the more numerical direction based on measured results rather than learned physics-based approximations because I think this approach is more tractable and easier to compute.

[tpr0p]

see #30 for cases where a single wavefront calibration at a point is not enough to get good hologram generation over FOV

[ichristen]

Thomas and I discussed in person the roadmap to resolve this issue. We're keeping this issue open to remind us to put all the pieces together to fully calibrate a field of view.