TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Three-eyed forehead in Stable Diffusion

112 pointsby hexomancerover 2 years ago

17 comments

A_D_E_P_Tover 2 years ago
This reminds me of something a few friends and I tried a couple of months ago: No matter what prompt was used, neither Midjourney nor Dream Studio could generate an image of a man wearing a red suit jacket with a blue shirt. (We were trying for red suit + blue shirt + white tie... but even just the first two proved impossible.) Presumably the combination is so unusual as to run counter to the training data of the models. Likewise for a forehead with three eyes.
评论 #34245898 未加载
评论 #34243895 未加载
评论 #34244210 未加载
评论 #34250530 未加载
评论 #34245362 未加载
评论 #34246454 未加载
narratorover 2 years ago
On a similar note to Stable Diffusion refusing to put 3 eyes in the middle of a sci-fi character's forehead: I have been experimenting with GPT-3 rewriting some of my sci-fi stuff. It's really funny because it right away tries to steer the plot into the most cliche sci-fi storyline and characterization possible where all the characters are perfect almost superhero like action heroes capable of incredible feats of strength and agility. My characters have a lot of flaws, and aren't impressive in an action movie sort of way so GPT-3 winds up being almost totally unusable.
terminal_dover 2 years ago
This is achievable without copy/pasting eyes: If you're using the Automatic111 GUI, go to img2img -> inpaint, mask the area for one eye (on the forehead), enter prompt, and set padding = 0 and denoising accordingly (0.4 - 0.6 would be acceptable). Repeat for all three eyes. You can add practically anything to an image with inpainting, provided your prompt and padding is correct.
sophrocyneover 2 years ago
This is what Invoke&#x27;s Stable Diffusion canvas solves for.<p><a href="https:&#x2F;&#x2F;youtu.be&#x2F;RwVGDGc6-3o" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;RwVGDGc6-3o</a>
andybakover 2 years ago
Dall-E has an interesting take on the problem: <a href="https:&#x2F;&#x2F;labs.openai.com&#x2F;sc&#x2F;JZIuAmvnELh8cMnBsLRVo5qk" rel="nofollow">https:&#x2F;&#x2F;labs.openai.com&#x2F;sc&#x2F;JZIuAmvnELh8cMnBsLRVo5qk</a>
TheDudeManover 2 years ago
Eyes? I thought they were rings&#x2F;piercings (in the original).
dwaltripover 2 years ago
One of the linked resources in the article is a great high-level overview of how Stable Diffusion works:<p><a href="https:&#x2F;&#x2F;stable-diffusion-art.com&#x2F;how-stable-diffusion-work&#x2F;" rel="nofollow">https:&#x2F;&#x2F;stable-diffusion-art.com&#x2F;how-stable-diffusion-work&#x2F;</a><p>It’s a quick read and I found it very helpful.
评论 #34252959 未加载
dangover 2 years ago
Recent and related:<p><i>Remaking old computer graphics with AI image generation</i> - <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=34212564" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=34212564</a> - Jan 2023 (73 comments)
Kaibeezyover 2 years ago
Not really on his forehead.<p>Six fingers, yeah, and also car wheels are a random mess of spokes and bolts.
评论 #34245346 未加载
nailloover 2 years ago
I only have a 3060 laptop gpu and an im2im run like this barely takes 3 seconds. It&#x27;s really fun and near real time if you keep the unet loaded in vram in between runs instead of re-loading it like calling a script would likely do.
boxedover 2 years ago
Square wheels is another fun example of how AI art is still super bad.
fortran77over 2 years ago
I’ve tried to get stable diffusion to draw 3-armed pianists, or pianists with extra fingers, and failed, probably for the same reasons this was difficult
评论 #34252258 未加载
Llamamoeover 2 years ago
In Midjourney this could be done using miltiprompting, and Automatic&#x27;s webui supports an analogous gesture with the AND keyword.
评论 #34246325 未加载
thrdbndndnover 2 years ago
Pretty cool article, I&#x27;d say the final result is kinda underwhelming, though.
评论 #34244937 未加载
stuaxoover 2 years ago
Disco diffusion is worth a go, it&#x27;s images are much more dream like.
greenhearthover 2 years ago
Looks like it stole Alex Ross style.
tiborsaasover 2 years ago
&quot;each inpainting took about 20 seconds which was quite annoying. But I could envision a future where generation is basically real-time, imagine navigating through possible generations using mouse wheel and tweaking the parameters and seeing the effects in real-time&quot;<p>This is really funny actually, considering what basic Photoshop tools are capable of out of the box :)
评论 #34245171 未加载
评论 #34245006 未加载
评论 #34244833 未加载
评论 #34244920 未加载
评论 #34245233 未加载