This is just img2img where first image with correct structure was generated by code.
Using “underdrawings” for accurate text and numbers
11–20 of 140 posts
Re: Using “underdrawings” for accurate text and numbers
#12Inpainting/guiding from a sketch is how I've always used diffusion models. I thought everyone did that, or at least everyone who wasn't just trying to get some arbitrary filler material without much care of what the output looked like.
Re: Using “underdrawings” for accurate text and numbers
#13This is just img2img where first image with correct structure was generated by code.
Pretty much what the author said- just gave some context for the uninitiated
Re: Using “underdrawings” for accurate text and numbers
#14This hack definitely falls in the “duh, why didn’t I think of that” category of tricks, but glad to now have it next time imagegen comes up short
Even the original stable diffusion app had image 2 image. It just didn’t work as well. I‘m not sure why this is supposed to be novel.
Re: Using “underdrawings” for accurate text and numbers
#15Transformers are great translators. So, yeah, starting with structured output like SVG is probably the best way to start.
It should be fairly trivial to fix any logic errors in the structured output, too.
Re: Using “underdrawings” for accurate text and numbers
#16This hack definitely falls in the “duh, why didn’t I think of that” category of tricks, but glad to now have it next time imagegen comes up short
Even the original stable diffusion app had image 2 image. It just didn’t work as well. I‘m not sure why this is supposed to be novel.
It's not novel in the sense that nobody knew about img2img. It's novel in the sense that nobody thought of using img2img to solve this problem in this way.
Re: Using “underdrawings” for accurate text and numbers
#17This is just img2img where first image with correct structure was generated by code.
Right, but you can use a different (codegen) model to make that code.
Re: Using “underdrawings” for accurate text and numbers
#18I wish the opposite was true: that when I tell Gemini I want "a diagram of X" that it immediately breaks out Python and mathplotlib, instead of wasting my time with Nano Banana.
Re: Using “underdrawings” for accurate text and numbers
#19This hack definitely falls in the “duh, why didn’t I think of that” category of tricks, but glad to now have it next time imagegen comes up short
Even the original stable diffusion app had image 2 image. It just didn’t work as well. I‘m not sure why this is supposed to be novel.
It’s obviously not a new model capability. But using this well-known, existing capability to solve this particular issue is only obvious after the fact.
It’s a useful trick to have in one’s toolbox, and I’m grateful to the author for sharing it.
Re: Using “underdrawings” for accurate text and numbers
#20I'm glad that we're making progress towards a deeper understanding of what LLMs are inherently good at and what they're inherently bad at (not to say incapable of doing, but stuff that is less likely to work due to fundamental limitations). There's similarity here with, for example, defining the architecture of software, but letting an LLM write the functions. Or asking an LLM to write you the SQL query for your data…
> There's similarity here with, for example, defining the architecture of software, but letting an LLM write the functions.
Not so long ago, this was how early adopters of LLM coding assistants claimed was the right way to use them in coding tasks: prompt to draft the outline, and then prompt to implement each function. There were even a few posts in HN on blogposts showing off this approach with terms inspired in animation work.