AnimeSuki Forums

Register Forum Rules FAQ Members List Social Groups Search Today's Posts Mark Forums Read

Go Back   AnimeSuki Forum > Anime Discussion > Older Series > Code Geass

Notices

Reply
 
Thread Tools
Old 2016-12-01, 16:10   Link #18321
Blackmambauk
Member
 
 
Join Date: Aug 2013
Location: England that is all
Code Geass Colored Memories Fic cover

http://jit-art.deviantart.com/art/Co...-Art-648143911

Just posting the Fic cover for Code Geass Colored Memories here for peeps to see.

Full credit to Jit Art for designing it, he did a splendid job on it and it was a great honour to work with him in the creation of it, the design of the top right figure was his own idea and one I am very impressed of.

As to whom the figure is in the top right corner. Well you will have to wait and see.

Here is the link to my fic here at ff.net.

https://www.fanfiction.net/s/1222886...lored-Memories

It is a companion fic to a fic called Code Geass Colorless Memories, which features Rai from Lost Colors as the main character.

Colored Memories focus's on Kaguya, Suzaku and Kallen as it's pov, tying into Colorless Memories at points, but also exploring the trio and others in ways CM does not.

https://www.fanfiction.net/s/1124204...Memories-Redux

While it starts out slowly and sticks a bit too close to Lost Colors and the Anime stuff at first.

It slowly create its own path and has a potentially really interesting Myth Arc to it.

Blackmambauk

Last edited by Blackmambauk; 2016-12-01 at 16:10. Reason: image link wouldn't work so changing to url link instead
Blackmambauk is offline   Reply With Quote
Old 2018-02-28, 15:36   Link #18322
EroKing
Senior Member
 
 
Join Date: Dec 2010
Age: 38










__________________

Last edited by EroKing; 2023-11-23 at 08:09.
EroKing is offline   Reply With Quote
Old 2018-03-24, 04:43   Link #18323
darthfury78
Senior Member
 
Join Date: Aug 2007
darthfury78 is offline   Reply With Quote
Old 2018-03-24, 04:45   Link #18324
darthfury78
Senior Member
 
Join Date: Aug 2007






Last edited by darthfury78; 2018-04-12 at 03:23. Reason: More Pictures found...
darthfury78 is offline   Reply With Quote
Old 2018-06-26, 14:48   Link #18325
EroKing
Senior Member
 
 
Join Date: Dec 2010
Age: 38


__________________

Last edited by EroKing; 2023-11-22 at 13:40.
EroKing is offline   Reply With Quote
Old 2021-03-01, 02:47   Link #18326
darthfury78
Senior Member
 
Join Date: Aug 2007
Spoiler:

Last edited by darthfury78; 2021-03-30 at 00:44.
darthfury78 is offline   Reply With Quote
Old 2022-01-21, 13:35   Link #18327
EroKing
Senior Member
 
 
Join Date: Dec 2010
Age: 38








__________________

Last edited by EroKing; 2023-11-20 at 07:03.
EroKing is offline   Reply With Quote
Old 2024-01-23, 08:48   Link #18328
Renegade334
Sleepy Lurker
*Graphic Designer
 
 
Join Date: Jul 2006
Location: Nun'yabiznehz
Age: 38
So...I hope you'll pardon the thread necromancy, but I figured I could just post these here as they are exclusively Code Geass pictures and this sub-forum got defibrillated back to consciousness a couple weeks ago with the Rozé of the Recapture trailer.

But...a disclaimer first?
These are all AI-generated by yours truly.

I'm aware of the controversy surrounding such software and how it can cheapen the hard work and creativity of blood-and-bone artists, but I promise this is a one-time thing. For those interested, I have appended the generation prompts and settings. All assets (models, LoRAs, upscalers, textual inversion embeddings) can be found at Civitai.

Click on every image to see their full version


TL;DR…
Prompts
Sorry; dynamic content not loaded. Reload?

This one here has perfect LoRA outputs (the characters are distinct from one another and do not suffer from accidental dimorphism), but the background (namely the window) shows a skewed perspective that is not shared by both Kallen and Milly. Bah, let's just pretend they're standing on an out-of-frame stairway, shall we?

TL;DR…
Prompt
Sorry; dynamic content not loaded. Reload?

So here we come to the reason why I'm doing this post at all: I'm celebrating a milestone that I've been trying to reach for a couple of months now: creating an output image with 1°) TWO different LoRA characters whose designs don't bleed into one another and 2°) successfully making Hires.fix upscale the base image, 3°) physically isolating the characters with Regional Prompter, 4°) all of that without crashing Stable Diffusion, consarnit. Condition #4 was the hardest of them to fulfill, as Hires.fix liked to crash before I started fiddling with venv startup parameters. #3 took me a lot of time to properly master and was the true secret to prevent Stable Diffusion from mixing the character models in a digital blender. I will spare you the endless cortege of eldritch...things SD produced before this.

Base image, in 768x512px. Note the blurriness (esp. the eyes) and note Kallen's right (left side for us) elbow: a blurry hand wrapped around it. Also take note of her waist right beneath it - it'll be important for the next picture.

TL;DR…
Prompt
Sorry; dynamic content not loaded. Reload?

Same image as above, but processed through Hires.fix and upscaled to 1536x1024px, showcasing how Stable Diffusion adds details and smoothens flaws and other artifacts, even correcting contextual errors. Look at Kallen's right elbow: the hand has now become a textile fold in her sleeve - and her waist has been altered to show that missing hand! Also, Milly's smile has been improved, as have the eyes, too. The decorative patterns across the glass windows have been altered and given better definition. Very pleasant result. But it took me a lot of time to get this.

Same prompt as above, of course.

NSFW
NSFW - avast, ye prudes! Thou hath been warned!
Sorry; dynamic content not loaded. Reload?

"But, Renegade - why just Milly, Kallen and Lelouch? Why not others?"
The reason is simple: their character LoRAs (character skinning models, if you will) were the best-looking and most reliable ones I've found so far. It's a bit too easy to find character LoRAs on Civitai, Pixai or Huggingface that seem more fit for hentai material (e.g. gigantic breast sizes) than normal/vanilla scenes. I recently tried a LoRA for Viletta Nu that only spat oversaturated and blurry stuff - thus earmarking it for deletion; my Lelouch LoRAs also tend to be rather temperamental, often spitting out illustrations whose subjects are barely 75% Lelouch-like or feel like they belong in a different anime from a different era (like Amuro Ray). The mileage also varies depending on the diffusion model used and any (ambiance|style|detailing) LoRAs you pair together. There is a LOT of trial and error involved in this. I might try other characters like C.C., or others from different series, though. Someday, though; not today.

Before you leave, I bid you remember one fact: for every beautiful or, should I say "seemingly perfect", picture the AI generator produces, there are quite likely dozens of very unsatisfying or outright failed ones hiding in the background or immediately banished to your operating systems' recycle bin. What you see posted here are essentially...the happy accidents - I honestly have no other way to describe them. It's just that - beautiful coincidences where the stars aligned perfectly. There is a whole art to harnessing roughly guiding Stable Diffusion's in-built randomness and making the best of it. There is a stupendous amount of trial and error involved in image generation (finding the right settings, the right prompt, the right negative prompt and the right assets!) and, given that I'm quite a bit short on free time nowadays, I'll be putting Stable Diffusion and both its wonders and horrors on the backburner. So...please: don't forward me requests like "can you generate a pic for me that shows yadda yadda yadda?". It's very time-consuming (especially for my aging PC) and sometimes even frustrating. But when you hit that rare unicorn? Yeah, makes it almost worth all the effort. The rest of the time, your prompt is more likely to accidentally summon Nyarlathotep or Y'golonac through your screen.
__________________
<< -- Click to enter my (dead) GFX thread.

Last edited by Renegade334; 2024-03-10 at 07:03. Reason: Matching prompts with the right outputs - sorry for the mix-ups
Renegade334 is offline   Reply With Quote
Old 2024-01-24, 21:48   Link #18329
Renegade334
Sleepy Lurker
*Graphic Designer
 
 
Join Date: Jul 2006
Location: Nun'yabiznehz
Age: 38
Okay, I lied. Somewhat...or is it just stretching the truth, though? Just think of this as a mere continuation of my previous post: following my recent breakthrough with Stable Diffusion's Regional Prompter extension and the Hires.fix upscaler, I tried different LoRAs, applying each of them on a single (NSFW) scene/setting. Each image was generated using the same, basic prompt, but each time adapted to reflect the subject character's design quirks (hair length and style; eye, skin and hair color; chest size, possible accessories; facial expressions) - unfortunately SD didn't obey all orders (some characters should be sitting but are actually standing) and in certain cases I even had to overly emphasize certain traits to get more accurate results (Rakshata Chawla's skin tone, which I had to correct afterwards using Photoshop, in the interest of full disclosure; I also had to add weights to Villetta's skin color, too).

NSFW
NSFW, AI pics - batch 2! (1024x1536px)
Sorry; dynamic content not loaded. Reload?
You'll note how difficult it is to generate pics that follow CLAMP's/Takahiro Kimura's style to a T or come even close to it. Some LoRAs fare better, others worse - your mileage will also vary according to the diffusion model used (Yesmix, meinamix, aamAnyloraAnimeMixAnime, etc.). I understand that better results can be gotten through networks but that's just extra workload for a program that is already crash-prone when upscaling to certain image sizes and is working with...not the most amazing amount of VRAM around, to be honest. I found more LoRAs to play with (Monica, Dorothea, Nunnally and Marianne), but at this moment I have not yet decided whether I'll generate something...but if I ever do, just check this post for edits and new additions.

EDIT:


TL;DR…
Prompts
Sorry; dynamic content not loaded. Reload?
Note: some LoRAs (Low-Rank Adaptation assets, essentially skinning files) are trained on visual material so specific it actually constrains the range of available results. The Marianne vi Britannia LoRA, for one, was extremely insistent on placing her indoors, most of the time inside a wooden corridor. I wanted her either in a garden or inside a palace. You can see in the prompts just how much weight (in the form of parentheses, which indicates how much importance the software must assign to these criteria) I had to give to "outdoors" and "French gardens", before the LoRA and the diffusion model finally deigned to indulge me and move the character to an outdoors setting. Same-o for the Julius Kingsley LoRA: absolutely impossible to get a side profile of Julius, only him facing the viewer - then I decided to add some chaotic/style/atmosphere LoRAs to experiment a little and vent my frustration. In both cases, the upscalers' eldritch-archwizard-magic did the rest (you can see the mind-boggling difference in detail between the upscaled, final versions and their original starting points: #1 and #2). Who the hell ever said AI image generation was a walk in the park? Certainly not for those who have a very clear idea of what the output should be compared to what you actually are served with.

Feel free to use the images as you please (save commercial purposes, of course). Toodles.
__________________
<< -- Click to enter my (dead) GFX thread.

Last edited by Renegade334; 2024-03-05 at 03:40. Reason: [UPDATED 2024/01/29: +1 image]
Renegade334 is offline   Reply With Quote
Old 2024-02-22, 00:31   Link #18330
darthfury78
Senior Member
 
Join Date: Aug 2007
Quote:
Originally Posted by Renegade334 View Post
Okay, I lied. Somewhat...or is it just stretching the truth, though? Just think of this as a mere continuation of my previous post: following my recent breakthrough with Stable Diffusion's Regional Prompter extension and the Hires.fix upscaler, I tried different LoRAs, applying each of them on a single (NSFW) scene/setting. Each image was generated using the same, basic prompt, but each time adapted to reflect the subject character's design quirks (hair length and style; eye, skin and hair color; chest size, possible accessories; facial expressions) - unfortunately SD didn't obey all orders (some characters should be sitting but are actually standing) and in certain cases I even had to overly emphasize certain traits to get more accurate results (Rakshata Chawla's skin tone, which I had to correct afterwards using Photoshop, in the interest of full disclosure; I also had to add weights to Villetta's skin color, too).

TL;DR…
NSFW, AI pics - batch 2! (1024x1536px)
Sorry; dynamic content not loaded. Reload?
You'll note how difficult it is to generate pics that follow CLAMP's/Takahiro Kimura's style to a T or come even close to it. Some LoRAs fare better, others worse - your mileage will also vary according to the diffusion model used (Yesmix, meinamix, aamAnyloraAnimeMixAnime, etc.). I understand that better results can be gotten through networks but that's just extra workload for a program that is already crash-prone when upscaling to certain image sizes and is working with...not the most amazing amount of VRAM around, to be honest. I found more LoRAs to play with (Monica, Dorothea, Nunnally and Marianne), but at this moment I have not yet decided whether I'll generate something...but if I ever do, just check this post for edits and new additions.

EDIT:


TL;DR…
Prompts
Sorry; dynamic content not loaded. Reload?
Note: some LoRAs (Low-Rank Adaptation assets, essentially skinning files) are trained on visual material so specific it actually constrains the range of available results. The Marianne vi Britannia LoRA, for one, was extremely insistent on placing her indoors, most of the time inside a wooden corridor. I wanted her either in a garden or inside a palace. You can see in the prompts just how much weight (in the form of parentheses, which indicates how much importance the software must assign to these criteria) I had to give to "outdoors" and "French gardens", before the LoRA and the diffusion model finally deigned to indulge me and move the character to an outdoors setting. Same-o for the Julius Kingsley LoRA: absolutely impossible to get a side profile of Julius, only him facing the viewer - then I decided to add some chaotic/style/atmosphere LoRAs to experiment a little and vent my frustration. In both cases, the upscalers' eldritch-archwizard-magic did the rest (you can see the mind-boggling difference in detail between the upscaled, final versions and their original starting points: #1 and #2). Who the hell ever said AI image generation was a walk in the park? Certainly not for those who have a very clear idea of what the output should be compared to what you actually are served with.

Feel free to use the images as you please (save commercial purposes, of course). Toodles.
Excellent pictures.
darthfury78 is offline   Reply With Quote
Old 2024-02-22, 01:31   Link #18331
Renegade334
Sleepy Lurker
*Graphic Designer
 
 
Join Date: Jul 2006
Location: Nun'yabiznehz
Age: 38
Thanks for the compliment; it wasn't easy and Takahiro Kimura/CLAMP's art style is kinda tricky to properly emulate. I do have more of these in my personal GFX thread, though - and I just appended the last two pics of the whole collection yesterday (I've decided to take a long break from AI; it's just too time- and resource-consuming for my tastes).
__________________
<< -- Click to enter my (dead) GFX thread.
Renegade334 is offline   Reply With Quote
Old 2024-03-01, 23:01   Link #18332
GundamFan0083
Senior Member
 
 
Join Date: May 2009
Location: classified
Quote:
Originally Posted by Renegade334 View Post
So...I hope you'll pardon the thread necromancy, but I figured I could just post these here as they are exclusively Code Geass pictures and this sub-forum got defibrillated back to consciousness a couple weeks ago with the Rozé of the Recapture trailer.

But...a disclaimer first?
These are all AI-generated by yours truly.

I'm aware of the controversy surrounding such software and how it can cheapen the hard work and creativity of blood-and-bone artists, but I promise this is a one-time thing. For those interested, I have appended the generation prompts and settings. All assets (models, LoRAs, upscalers, textual inversion embeddings) can be found at Civitai.
As an artist (you can find my work here: https://www.deviantart.com/kodai-okuda), I for one will say that I've no problem with AI art. Love it actually and I'm learning how to use it myself because it is fast, efficient, and cost-saving in terms of time and money.
So no worries (at least from me).
You did a fantastic job with the prompts.
Kudos for getting the model to generate such accurate and amazing Code Geass artwork.
Very nice.
__________________
GundamFan0083 is offline   Reply With Quote
Old 2024-03-02, 03:49   Link #18333
Renegade334
Sleepy Lurker
*Graphic Designer
 
 
Join Date: Jul 2006
Location: Nun'yabiznehz
Age: 38
Thank you for your kind words, but, alas, that sentiment isn't universally shared. As a concrete example, Imgur loves to downvote to oblivion Community posts with the tags "aiart" and/or "stablediffusion"; they're a lot more vocal in their rejection of AI artistry. Real, flesh-and-blood artists could also feel discouraged by this new form of competition; as an occasional Photoshopper, I, for one, know that I can never match this kind of graphical prowess (not to mention, I've grown somewhat rusty in that field - it's been a while since I made user signatures for forums). However, I've found that SD still requires a lot of Photoshop post-processing, especially to fix some of the finer stuff like hands and faces - and sometimes the eyes.

For the prompts, I stopped appending them because I've started making increasingly more complex pictures, leveraging regional prompts (I've switched from the time-honored Regional Prompter to Multidiffusion Upscaler for Automatic 1111, which also has a regional prompt option but with a far better boundary box-adjustment system), Hires fix upscaling, ControlNet reference models and daisy-chained After Detailer tasks with fine-grain instructions. These combinations make for very long text outputs, which quickly degrade the readability of the forum's post editor.

Example:
TL;DR…
eye-watering prompt
Sorry; dynamic content not loaded. Reload?
...Now imagine five or six more of this, stacked together. Yikes.

Anyway, I might as well repost my pics from my thread here since they're all Code Geass-related. Without all the tedious, technical commentary, of course.

NSFW
***Some*** of these are not safe for work
Sorry; dynamic content not loaded. Reload?
NSFW
same images as above, with Photoshop filters and sharpening
Sorry; dynamic content not loaded. Reload?

That'll be all for a while, as I've started my long AI hiatus.

EDIT - Imgur really loves to delete those two sets of pictures (original and Photoshop-filtered), so I'm uploading them on a different hoster (Imgbox). Sorry for the inconvenience:
NSFW
nsfw
Sorry; dynamic content not loaded. Reload?
__________________
<< -- Click to enter my (dead) GFX thread.

Last edited by Renegade334; 2024-03-09 at 09:08.
Renegade334 is offline   Reply With Quote
Old 2024-03-02, 16:21   Link #18334
GundamFan0083
Senior Member
 
 
Join Date: May 2009
Location: classified
Quote:
Originally Posted by Renegade334 View Post
Thank you for your kind words, but, alas, that sentiment isn't universally shared. As a concrete example, Imgur loves to downvote to oblivion Community posts with the tags "aiart" and/or "stablediffusion"; they're a lot more vocal in their rejection of AI artistry. Real, flesh-and-blood artists could also feel discouraged by this new form of competition; as an occasional Photoshopper, I, for one, know that I can never match this kind of graphical prowess (not to mention, I've grown somewhat rusty in that field - it's been a while since I made user signatures for forums). However, I've found that SD still requires a lot of Photoshop post-processing, especially to fix some of the finer stuff like hands and faces - and sometimes the eyes.

For the prompts, I stopped appending them because I've started making increasingly more complex pictures, leveraging regional prompts (I've switched from the time-honored Regional Prompter to Multidiffusion Upscaler for Automatic 1111, which also has a regional prompt option but with a far better boundary box-adjustment system), Hires fix upscaling, ControlNet reference models and daisy-chained After Detailer tasks with fine-grain instructions. These combinations make for very long text outputs, which quickly degrade the readability of the forum's post editor.
Prompts themselves are becoming a skill with which to create the desired result. It is not an easy thing to do. Not too mention that most AI Art engines will generate a larger number of images that are "mistakes" as they do "good" images that are useful. I'm not as advanced in it as you are, but I'm learning and it is the learning that is likely what is causing many traditional artists to oppose AI Art.

I've done hand drawn artwork myself (here is an example of an original Code Geass piece):

Spoiler for Large image:


I've since moved on to 3D digital artwork for my own science fiction.

Spoiler for Large Image:


However, I'm not afraid to use AI for that science fiction because I don't do character images well at all and employing an artist to design and create book covers or interior artwork is cost-prohibitive. I'm working on a TTRPG series of books right now for my Nexus Arcana series of traditional novels and the estimated cost to hire artists to do the interior artwork was upwards of $50,000 US. That's not possible for me to afford or justify. AI however (with commercial license) makes it possible at a greatly reduced cost and that's why we're seeing every major publisher of books going over to it. Video game companies are also going to it for backgrounds, and other art that doesn't require the skill of a flesh-and-blood artist. Traditional artists will still be needed for the concept art, detailed art, etc. because as you know, getting AI to actually craft something like a whole anime series is not possible, yet.

My own AI art is akin to this (nothing spectacular, but functional for its purpose).

Book 1 is purely AI art:


Book 2 is a combination of AI art with a photo background completed in photoshop:


Book 3 is a combination of AI art with original 3D graphic models composited in photoshop into one image:


It will take some time, but once traditional artists figure out that AI can cut production time and increase workflow significantly for backgrounds, and other uses to improve their original art, methinks the resistance to the technology will subside considerably (there will always be hold outs).
__________________
GundamFan0083 is offline   Reply With Quote
Old 2024-03-03, 19:40   Link #18335
Renegade334
Sleepy Lurker
*Graphic Designer
 
 
Join Date: Jul 2006
Location: Nun'yabiznehz
Age: 38
TL;DR…
AI discussion, off-topic
Sorry; dynamic content not loaded. Reload?

Aaaanyway, here's one of the few pics left at the bottom of my figurative drawer. While unburdening my HDD from now-unneeded SD stuff, I found a few base images I thought were worth further refining...but they unfortunately couldn't clear the upscaling stage in SD, because of low VRAM-induced software crashes. If I ever manage to clear the technical hurdles that caused those tasks to fail, I'll post the outputs here. Only a very small handful of those left, though (IIRC there should be at least one Jeremiah and one Shirley pic), so don't expect a deluge of new arrivals.







__________________
<< -- Click to enter my (dead) GFX thread.

Last edited by Renegade334; 2024-03-09 at 07:42. Reason: Added Snowy Shirley
Renegade334 is offline   Reply With Quote
Reply

Tags
images, wallpaper

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -5. The time now is 03:35.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
We use Silk.