Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: New AI edits images based on text instructions

github.com

Post reply on HN
Print view View original

231–240 of 241 posts

Re: Show HN: New AI edits images based on text instructions

#231

Post by johndough »

social_quotient wrote: post #127

Slightly off topic. I’ve been looking for an easier way to replace the text in these ai generated images. I found Facebook is working on it with their TextStyleBrush - https://ai.facebook.com/blog/ai-can-now-emulate-text-style-i... but have been unable to find something released or usable yet. Anyone aware of other efforts?

The authors of TextStyleBrush cite SRNet, which is available at https://github.com/youdao-ai/SRNet but probably has worse quality. I don't know of others, but I have not looked very hard either.

Re: Show HN: New AI edits images based on text instructions

#232

Post by treeman79 »

dr_dshiv wrote: post #230
treeman79 wrote: post #229

Earlier quoted context omitted.

Found out I couldn’t see in stereo. Got prism glasses. Completely blew my mind to seen ”depth” for the 1st time. Had no idea I couldn’t. Never had any trouble. Apparently without prism glasses my vision just switches from one eye to the next every 30 seconds. Completely seamlessly.

Wow! Any brand suggestions? And, are these the same prism glasses as those that let you watch tv laying in bed?

No. They are for double vision. Had a series of small strokes. Caused high cranial pressure which caused six nerve palsy. I take a pill now and it fixes double vision. I lose stereo that way. But prism glasses are very distracting to wear.

Re: Show HN: New AI edits images based on text instructions

#233

Post by CptanPanic »

GordonS wrote: post #63

What are the most affordable GPUs that will run this? (it said it needs CUDA, min 11GB VRAM, so I guess my relatively puny 4GB 570RX isn't going to cut it!)

Google colab free acount, you get access to 15GB vRAM T4 GPU, or Kaggle which gives you access to 2xT4's, or one P100 GPU.

Re: Show HN: New AI edits images based on text instructions

#234

Post by int_19h »

wongarsu wrote: post #171
babarock wrote: post #156

Earlier quoted context omitted.

Genuinely curious what type of content would be considered illegal here. - the tool is drawing original content. - the tool is executing on my laptop. - the output image is not shared with anyone. In a way, whatever this tool can do, MS Paint could (theoretically) do. Or am I misunderstanding the whole thing?

There's a great wikipedia page on this very topic [1]. In some countries like Germany fictional porn isn't considered porn, while other countries like Australia or France consider the possession of drawings of naked minors a crime worthy of jail sentences. And then there's the US where having the images on a computer is fine, as long as they aren't sent over the internet and the computer never crosses state lines. In…

When you stop and think about it, it is kind of insane that a fictional drawing in private possession, not even communicated to anyone, is already sufficient to land someone in prison in so many jurisdictions. Isn't this a clear case of legislating moral purity (as opposed to actual harm)?

Re: Show HN: New AI edits images based on text instructions

#235

Post by andrijeski »

sandworm101 wrote: post #145

Fireworks. These AI tools seem very good at replacing textures, less so about inserting objects. They can all "add fireworks" to a picture. They know what fireworks look like and diligently insert them into "sky" part of pictures. But they don't know that fireworks are large objects far away rather than small objects up close (see the Father Ted bit on that one). So they add tiny fireworks into pictures that don't ha…

Given the appropriate training and test set, we can build a model that can overcome these issues, right?

Re: Show HN: New AI edits images based on text instructions

#236

Post by pfd1986 »

bryced wrote: post #149
pfd1986 wrote: post #98

Super nice. Would this work if I have my own version of fine-tuned SD? Also, curious how / whether this is different from img2img released by SD. Thanks!

This is itself it's own finetuned version of SD so now it won't work with alternative versions. img2img works by just running normal stable diffusion img2img on a noised starting image. As such it destroys information at all parts of the image equally. This new model uses attention mechanisms to decide which parts of the image are important to modify. It can leave parts of the image untouched while making drastic cha…

Perfect explanation, thank you!

Re: Show HN: New AI edits images based on text instructions

#238

Post by googie »

googie wrote: post #237

How to make it use my GPU (I have RTX 3070)? It complains about using sloooow CPU, but I don't see option to switch to GPU, which I think should be sufficient...? I'm running it on Windows 10.

Note, that I have the CUDA installed. Still the imaginAIry runs on CPU :(

Re: Show HN: New AI edits images based on text instructions

#239

Post by zepearl »

Thanks a lot!!!

Works perfectly for me (Gentoo Linux + nVidia RTX3060 12GiB VRAM - I installed last week your package and it just worked, experimenting with it since then, telling about it parents & colleagues).

The results (especially in relation to "people's faces") can vary a lot between ok/scary/great (I still have to understand how the options/parameters work), all in all it's a great package that's easy to handle & use.

In general, if I don't specify a higher output resolution setting than the default (512x386 or something similar), with e.g. "-w 1024 -h 768", then faces get garbled/deformed like straight from a Stephen King novel => is this expected?

Cheers :)

Re: Show HN: New AI edits images based on text instructions

#240

Post by naderkhalil »

punkspider wrote: post #25
ilaksh wrote: post #6

Does anyone know if there is something like Google Cloud for GPUs but with an easy way to suspend the VM or container when not in use? Maybe I am just looking for container hosting with GPUs. I am just trying to avoid some of the basic VM admin stuff like creating, starting, stopping for SaaS if someone already has a way to do. Maybe this is something like what Elastic Beanstalk does.

I think there's https://brev.dev and https://banana.dev

Hey, founder of Brev.dev here. Brev lets you suspend the instances when not in use, and also auto-stops it after 3 hours of inactivity to avoid expensive surprises. Would love for you to give it a shot
Post reply on HN