bryced

8 ms·

31.
by bryced 4y ago
Works fine for me, you just need to adjust the strength of the edit.
32.
by bryced 4y ago
It is stable diffusion but yes my fork does not have the memory optimizations needed to run it on only 4gb
33.
Show HN: New AI edits images based on text instructions(github.com)
1098 points by bryced 4y ago|227 comments
34.
by bryced 4y ago
> Another interesting (albeit much harder) thing to try would be to take the image output from SD, run it through MiDaS to get a new depth map with additional details. Diff the depth maps in 3d space and update the geometry to match befo
35.
Stable Diffusion 2.1 – renders people again – (Python package for Mac and Linux)(github.com)
3 points by bryced 4y ago|1 comments
36.
by bryced 4y ago
Stable Diffusion 2.1 fixes the bug in 2.0 that caused people to be rendered poorly. Example of a fun way to make desktop wallpaper (make sure you set the desktop to tile the image): >> pip install imaginairy >> aimg imaginAIry&g
37.
by bryced 4y ago
Possibly. Haven't tried. In principle should be possible.
38.
by bryced 4y ago
I don't think it works with anaconda on any OS.
39.
by bryced 4y ago
You're probably noticing the newest sampler, which also works with 1.5.
40.
by bryced 4y ago
Yeah I tried to get it running but it kept crashing with "out-of-ram" errors. Good idea to version the notebook.
41.
by bryced 4y ago
- text-based masks use a clipseg model. - the boolean mask logic is unique to this library - the face fixing is done by CodeFormer
42.
by bryced 4y ago
All the same issues as migrating 1.5 to M1s. It went fast because I upgraded my existing codebase that had those fixes already instead of building of the new compvis one.
43.
by bryced 4y ago
It might work on windows but I haven't tested it there.
44.
Stable Diffusion 2.0 on Mac and Linux via imaginAIry Python library(github.com)
234 points by bryced 4y ago|50 comments
45.
by bryced 4y ago
Try out the pre-release like this: `pip install imaginairy==6.0.0a0 --upgrade` New 512x512 model supported with all samplers and inpainting New 768x768 model supported with the DDIM sampler only Not yet supported is the upscaling and depth
46.
by bryced 4y ago
> Will AI lead to UBI? (56:09) > Maybe. It'll either lead to UBI and utopia or panopticon that we can never escape from because the models that were previously used to focus our attention and service ads will be used to control o
47.
Q&A with Emad Mostaque of Stable Diffusion. Formatted Transcript with Outline(github.com)
10 points by bryced 4y ago|2 comments
48.
by bryced 4y ago
The Q&A was packed full of information. I wanted to make this more accessible if you don't have time to listen to a long recording. Created with Whisper and a few hours manually editing the text.
49.
Q&A with Emad Mostaque (of Stability.ai, Stable Diffusion creator) – 2022-10-10(drive.google.com)
4 points by bryced 4y ago|2 comments
50.
by bryced 4y ago
Super fascinating Q&A with Emad about the future of AI and upcoming model releases.
51.
by bryced 4y ago
oh that's great - thank you!
52.
by bryced 4y ago
For code that doesn't have good pypi packaging I vendor it using scripts in the makefile
53.
by bryced 4y ago
Thanks for the feedback!
54.
by bryced 4y ago
the "alpha" is already built into the tooling and specified independent of the mask. IE. the stable diffusion inpainting takes hints from what you leave and "decides" what to keep
55.
by bryced 4y ago
This python library does the same thing (but didn't get traction when I posted it yesterday): https://github.com/brycedrennan/imaginAIry#automated-replace... https://news.ycombinator.com/item?id=32
56.
Txt2mask – automated object replacement in images (Stable Diffusion)(github.com)
20 points by bryced 4y ago|3 comments
57.
by bryced 4y ago
Really cool new feature I just integrated into this python library: AI object replacement. You just specify something like "replace the fruit with gold coins" and Bam! it happens. Makes editing way way easier. Have fun!
58.
Pythonic generation of Stable Diffusion images(github.com)
5 points by bryced 4y ago|1 comments
59.
by bryced 4y ago
For python developers with appropriate hardware, this should be the easiest way to get started generating images.
60.
by bryced 4y ago
Doesn't seem very mysterious. When taken in context it looks like an innocuous tweet but to a twitter censorship algorithm that isn't aware of the context it looks like a command for someone to die.

More ›