Non-iterative Learned Hologram Calculation

[kaanaksit]

We provided an overview of optimizing holograms using iterative methods. Iterative methods are computationally expensive and unsuitable for real-time hologram generation. We challenge our readers to derive a learned hologram generation method for multiplane images (not single-plane like in our example). This development could either rely on classical convolutional neural networks or blend with physical priors explained in this section. The resultant method could be part of odak.learn.wave submodule as a new class odak.learn.wave.learned_hologram. In addition, a unit test test/test_learn_hologram.py has to adopt this new class. To add these to odak, you can rely on the pull request feature on GitHub. You can also create a new engineering note for arbitrary surfaces in docs/notes/learned_hologram_generation.md.