VN Ren'Py Abandoned Kingdom of Deception [v0.14.2.2] [Hreinn Games]

[TheMagister]

Okay I attempted to train my own "Momo Style" LORA. (Using KOD / NN / Rule34)

It does not have any concepts of characters..
So there is no such thing as "Sabia" or "Neve" in the tagging, if you want those characters you have to describe them.

I did a few quick tests .. with the following rules:
- No Controlnets (meaning the poses are purely prompt based, each image may look slightly different)
- No image touchups or fixups or seed cherry picking
- No Segmenting (each image will have small variations, like clothing changes or background changes)

Known Issues So Far:
I can already see a problem with my tagging, I completely trashed "ponytail" with "folded ponytail" training. I will need to pick a single word tag that doesn't do that.


The image prompts left to right are:

[Describing Sabia as best I can]
[Describing Neve but with short hair and elf ears]
[Describing Neve but with short hair and cat ears]

The background is just "Sunny Day, Village, Huts, Grass"


Feedback appreciated - is the art style over cooked or under cooked?

I will likely need to do another run (or 10.. ), so if people think the style is wrong / bad - please say so.

- Focus criticism / feedbacl on the art style (both background & characters), the color palette, line work, shading style, etc.
- Ignore inconsistencies in height / clothing / pose. Those are fixed in a different way.

I would say Neve doesn't look fierce enough. Sabia as well, but I feel like with her it's less of a problem but Neve was already highly confident and competent, and much more take-charge and domineering.

 

[DreamingAway]

I would say Neve doesn't look fierce enough. Sabia as well, but I feel like with her it's less of a problem but Neve was already highly confident and competent, and much more take-charge and domineering.

I think that's just me being lazy.
I added "confident, serious expression", is this better or still not hitting home?

If it's still not right can you show me an example?

 

[Darcel Marvis]

Okay I attempted to train my own "Momo Style" LORA. (Using KOD / NN / Rule34)

It does not have any concepts of characters..
So there is no such thing as "Sabia" or "Neve" in the tagging, if you want those characters you have to describe them.

I did a few quick tests .. with the following rules:
- No Controlnets (meaning the poses are purely prompt based, each image may look slightly different)
- No image touchups or fixups or seed cherry picking
- No Segmenting (each image will have small variations, like clothing changes or background changes)

Known Issues So Far:
I can already see a problem with my tagging, I completely trashed "ponytail" with "folded ponytail" training. I will need to pick a single word tag that doesn't do that.


The image prompts left to right are:

[Describing Sabia as best I can]
[Describing Neve but with short hair and elf ears]
[Describing Neve but with short hair and cat ears]

The background is just "Sunny Day, Village, Huts, Grass"


Feedback appreciated - is the art style over cooked or under cooked?

I will likely need to do another run (or 10.. ), so if people think the style is wrong / bad - please say so.

- Focus criticism / feedbacl on the art style (both background & characters), the color palette, line work, shading style, etc.
- Ignore inconsistencies in height / clothing / pose. Those are fixed in a different way.

View attachment 3068253

You can also use the gallerys on e-hentai.org for the training of the Lora, you can pretty much find the entire KoD gallery there.

 

[scheriel]

I think you guys produced more content in 1 day than Hreinn in roughly 8 months

 

[Great730]

I think these 3 arts can be prompted by sabia's domineering mother

Kira's sadistic and dominating nature you are so cute I think you deserve another blow

And the evil face of Sister Sabia, I hope it will help

 

[DreamingAway]

Just experimenting with creating Map Icons / Map Terrain. (I generated several images then cut out things I liked)
None of this was actually in my dataset though..

 

[Jinsoyun]

I'll have to read all this AI generating discussion once i have the time, because i have no idea about it's technicalities.

But in the meanwhile, may i just mention that the simple imitation of Nomo's art is exciting enough for us here to completely overshadow the new release. This is wild.

 

[Negativpro0]

View attachment 3067421 View attachment 3067422

Hands aside i think this looks great, though fairly, Nomo is not good at drawing hands either. It feels like lot of progress is being made here, i look forward to where we get.

 

[WYKYDTR0N]

AI shills begone! That garbage always looks the same; same lighting, same shading, same faces, same proportions. Eyes are wrong, lips are wrong, expressions are bad. It looks like the heartless, soulless slop that it is.

I know the game is fucked and the devs are either lazy or greedy or both, but let's treat the corpse with a little dignity instead of stringing it up like a puppet and making it dance. I know the art has gotten worse (somehow) but even at its worst it's still actually ART instead of just an image.

 

[DreamingAway]

This is not img2img.

I used the left one to generate a openpose rig - then used that rig to generate the right side images..
For those who don't use SD, the left image itself is not used at all to make the right one(s), I only used the left image to generate a skeleton rig which is basically a JSON file of lines that represent bones / fingers that can be used to pose characters, the pixels / original image is then thrown away, it's just a way to copy the pose, not the actual aesthetic.

Noxian Nights sequel?


There are small fixes / clean ups that could be done on the below images..
I was rushing a little because I'm trying to test how well the LORA works..

The goal should be:
- very consistent poses / facial expressions without relying on a crutch like img2img, because ideally you want to create multiple clothing types
- have very deterministic outcomes when you prompt or pose something, you don't want "randomness" coming through your LORA / Model otherwise it makes it very difficult to create assets.


.

 

[BigBiggieBig]

This is not img2img.

I used the left one to generate a openpose rig - then used that rig to generate the right side images..
For those who don't use SD, the left image itself is not used at all to make the right one(s), I only used the left image to generate a skeleton rig which is basically a JSON file of lines that represent bones / fingers that can be used to pose characters, the pixels / original image is then thrown away, it's just a way to copy the pose, not the actual aesthetic.

Noxian Nights sequel?


There are small fixes / clean ups that could be done on the below images..
I was rushing a little because I'm trying to test how well the LORA works..

The goal should be:
- very consistent poses / facial expressions without relying on a crutch like img2img, because ideally you want to create multiple clothing types
- have very deterministic outcomes when you prompt or pose something, you don't want "randomness" coming through your LORA / Model otherwise it makes it very difficult to create assets.


View attachment 3069262 .

yo....

 

[xhyde]

This is not img2img.

I used the left one to generate a openpose rig - then used that rig to generate the right side images..
For those who don't use SD, the left image itself is not used at all to make the right one(s), I only used the left image to generate a skeleton rig which is basically a JSON file of lines that represent bones / fingers that can be used to pose characters, the pixels / original image is then thrown away, it's just a way to copy the pose, not the actual aesthetic.

Noxian Nights sequel?


There are small fixes / clean ups that could be done on the below images..
I was rushing a little because I'm trying to test how well the LORA works..

The goal should be:
- very consistent poses / facial expressions without relying on a crutch like img2img, because ideally you want to create multiple clothing types
- have very deterministic outcomes when you prompt or pose something, you don't want "randomness" coming through your LORA / Model otherwise it makes it very difficult to create assets.


View attachment 3069262 .

Sir, im gonna need you to post that LORA pls.

 

[TERES]

New update KoD 0.14.1

 

[Partyboy78]

The last few pages have reminded me that once again, just like Viressa's add-ons, the community seems to have a better chance of actually keeping this game alive & interesting.

 

[DreamingAway]

Sir, im gonna need you to post that LORA pls.

That's the plan.
I'll have it up in a few "Nomo-Months".... (so.. 10 years.. )

--

Just kidding ~ ~

 

[Enigmanic]

If the Ai thing actually worked Hreinn will go broken

Not necessarily. Nomo could make things a lot easier on himself by using AI.
step 1 - He could have new images generated by AI using his art style.
step 2 - Next, he can load the images into whatever art program he uses. Then he can erase and redraw parts of the image to correct errors and bring it in line with his vision for the scene.
step 3 - Profit!

Of course, he still needs NDS to do his fucking job. Plus Monsinne needs to develop actual talent and stop writing like he's a 12yo who just discovered masturbation.

 

[DreamingAway]

I wouldn't suggest plagiarizing KoD. Instead I would write an original story with a similar theme and include characters similar to, if not inspired by, the principal cast of KoD. Then we can use AI to imitate that art style. It wouldn't be difficult at all to draft a similar story.

Is it too late to write Noxian Nights 2 instead?
Can we include Akali and Fiora?

 

[Enigmanic]

Is it too late to write Noxian Nights 2 instead?
Can we include Akali and Fiora?

If that's more your cup of tea. Though I wouldn't use existing characters. I could parody versions of those two characters after learning a bit about them. But no way am I touching Overwatch or LoL. I can't stand those games.

 

[Deleted member 4951110]

This is not img2img.

I used the left one to generate a openpose rig - then used that rig to generate the right side images..
For those who don't use SD, the left image itself is not used at all to make the right one(s), I only used the left image to generate a skeleton rig which is basically a JSON file of lines that represent bones / fingers that can be used to pose characters, the pixels / original image is then thrown away, it's just a way to copy the pose, not the actual aesthetic.

Noxian Nights sequel?


There are small fixes / clean ups that could be done on the below images..
I was rushing a little because I'm trying to test how well the LORA works..

The goal should be:
- very consistent poses / facial expressions without relying on a crutch like img2img, because ideally you want to create multiple clothing types
- have very deterministic outcomes when you prompt or pose something, you don't want "randomness" coming through your LORA / Model otherwise it makes it very difficult to create assets.


View attachment 3069262 .

oh my...

 

[Deleted member 4951110]

Is it too late to write Noxian Nights 2 instead?
Can we include Akali and Fiora?

make a discord or something, I feel like this has so much potential