[Findeton]

NeRFs are very cool and probably the future, but plain old multi-plane images (MPIs) still have their use case. If you want to create a real-time video light-field, MPIs are very fast to render. Although MPIs require a large amount of memory/RAM, rendering them is very fast. In comparison, NeRFs take very little space but rendering in real-time would be challenging at the moment as the NN is an implicit representation of the scene and you need to call the NN multiple times to render each pixel.

[fudged71]

Have you seen this optical neural network technique? It seems like one way you could get fast (real-time) results from a complex neural network https://www.photonics.com/AMP/AMP_Article.aspx?AID=63751

[Findeton]

That's very cool but right now that's not really viable is it? Each frame would require to 3d print a different NN?

[fudged71]

Yes I said “could” as a path to future utility. Light-speed complex computation.

[Findeton]

If they make that programmable somehow it could be a great advancement.