Closed bryanray closed 1 year ago
I dont use it, and im not a special snowflake so i don't imagine im the only one. I run an external image viewer that does a lot more than a built in gradio image browser is every going to do.
i know you are not a special snowflake, so are we all.
BUT many of us might use it as archiver, viewer(you dont say:)) to study different prompts and configs, you can send the cmd directly to the input area, and delete failed stuffs sometimes, it is good to keep them tho to remind whats bad...i have an output folder that's over 5gb, i use it extensively as i am a professional designer.
it is so useful: IF you have enabled this: --listen --enable-insecure-extension-access allowed yourself to access SD webui via your phone/ipad/laptop, as a remote control (ip http URL) THEN without a built-in or plugin image browser, how do you view your works? get it?
I mostly dont want to open tons of folders, webpages, it will cause cluster... although I have 3 monitors and a 4090 deskstop setup with a 3060 laptop on the side, it is so ridiculous to think this tool is just for Porn Generation, and Waifu for weebs.....
if you know how to squeeze the power out of this tool, then you will want to have certain professional utilities. UNLIKE IF all you need is just PORN Generation / Weebie Waifu / pervert picture THEN well, nothing really is matter, even more so cos you will wan to hide them :) just like incognito mode
cheers
I dont use it, and im not a special snowflake so i don't imagine im the only one. I run an external image viewer that does a lot more than a built in gradio image browser is every going to do.
Can it tell you the prompt and seed used in the creation of the image as well as other important metadata? Can you easily send the image to to img2img and txt2img along with this metadata to easily reproduce and fine tune results?
I dont use it, and im not a special snowflake so i don't imagine im the only one. I run an external image viewer that does a lot more than a built in gradio image browser is every going to do.
Can it tell you the prompt and seed used in the creation of the image as well as other important metadata? Can you easily send the image to to img2img and txt2img along with this metadata to easily reproduce and fine tune results?
no, but i can still drag into PNG info (or of course direct into wherever if i dont care about the metadata).
my point is, it's not everyone's use case to need this functionality built in, the point of modularity is being modular... the more things you add as core features, the more the whole thing slows down whether you want those features or not.
so you'd want to put in a toggle to disable the feature... but that then asks the question, why have an opt out, instead of the current opt in. is it really that hard to install an extension ?
the webui itself already HAS tons of what you would've described as "not useful to you" functions, i dont really mind them good functions.
yes modularity is all good, BUT you gotta WAIT for the features if they are MISSING, OR if the existing extensions are BROKEN, OR no-maintenance at all, out of date.....get it? THAT IS WHY built-in, is, the only sollution, cos it is in the loooop...
the point of this request is, as the development goes, there will be problems between dev, that is out of your scope of understanding why it is requested in the first place..... the image browser forker ... why forked? ......because what i have just said above..... https://github.com/AlUlkesh/stable-diffusion-webui-images-browser/issues/100#issue-1640872891
hey if I know how to code, I would have done it myself :) but i suck at it :)
I think another issue is stable-diffusion-webui-images-browser is unmaintained, and the most recently maintained version is a fork, but because Google doesn't index GitHub forks it still puts the unmaintained version at the top of the search results, which becomes confusing
For me I have eschewed any extensions to browse images and use hydrus instead with this script https://github.com/space-nuko/sd-webui-utilities/blob/master/import_to_hydrus.py
I think another issue is stable-diffusion-webui-images-browser is unmaintained, and the most recently maintained version is a fork, but because Google doesn't index GitHub forks it still puts the unmaintained version at the top of the search results, which becomes confusing
For me I have eschewed any extensions to browse images and use hydrus instead with this script https://github.com/space-nuko/sd-webui-utilities/blob/master/import_to_hydrus.py
no maintenance exactly, and google index the out of date version. thanks for the recommendation https://hydrusnetwork.github.io/hydrus/index.html kudos to those responsible adults who know real issue. in fact there are lib in python already doing good image browsing, maybe whats left for the sd dev team to do is import it and add png metadata functions and such.... is it possible for you guys?
This was previously part of the webui and later removed, so I'm going to go ahead and close this as not planned.
Is there an existing issue for this?
What would your feature do ?
i think the image browsing capability should be built in with stable diffusion, rather than a extension, it is such an essential part of the functionality.
After seeing so many failed from users who updates :)
it is just like img2img interrogation, it is built-in, beautifully, with interrogator models in local folder, this is good example against not so great lack of update plugins, like wd-tagger, whcih doesnt allows you to use local folder to store interrogator models, it must download and put in temp folder somewhere unspecified. GEEEz so ridiculous... another interrogator broken plugin is clip, a1111 has better option and built-in too.
so i bet a1111 should have make a built in image browser for good, it is so widely used.
different dev team will eventually has communication problems, so it is best to just make it built in, flush and clean.
Proposed workflow
omg
Additional information
No response