sign-language-processing/sign-language-processing.github.io

text-to-video: write about video diffusion models

Opened this issue · 2 comments

Write in general about text to video diffusion models, for example
https://twitter.com/MagusWazir/status/1640555696750993415

specify that they are not really understandable, and we can't tell where the error happens due to the end-to-end nature of them

signllm is an example of a terrible production system, but a nice ControlNet+LORA visualization. we already write about ControlNet, but not about AnimateDiff, LORAs, and whatever other methods they use to generate nice looking humans