Re:Real-time image editing using latent consistency models
by foota on Friday November 10, 2023 @08:56PM
(#38226883)
Would that motion vector have to include like a color delta?
[ Parent
| on HN ]
Re:Real-time image editing using latent consistency models
by billconan on Saturday November 11, 2023 @12:05AM
(#38227781)
if you are talking about latent diffusion, No, the "particle" is in a hyper-dimensional space, like for example a 10k-dimensional space. We are not supposed to interpret the meaning of that vector.
and when that particle has moved to the right location, there is a decoder that converts it into an image. The decoder network knows how to interpret it.
Re:Real-time image editing using latent consistency models
by dvrp on Saturday November 11, 2023 @02:14AM
(#38228219)
Exactly.
I actually wrote a micro essay on Twitter the other day about the meaning of the classic encoder decoder network. It’s beautiful.
But yeah!
-
For reference: https://x.com/asciidiego/status/1722544108252836119