• Sign up or login, and you'll have full access to opportunities of forum.

Redheads in Peril

Go to CruxDreams.com
mermaidhunter

mermaidhunter

Master Crucifixion Technician
Well, it began to seem to me that nobody was going to post some redheads on crosses, so I threw some together.

A redhead bound being marched out to her crucifixion site.
beautifulRedhead_001.jpg


She is thinking this is not as much fun as I said it would be.
RedheadinaMeadow_L001.jpg


She perked up when I promised her a beer afterwards, if she lives.
RedheadinaMeadow_L002.jpg


She is becoming concerned, thinking that she might not make it to the beer.
RedheadinaMeadow_L003L.jpg


She is struggling to make the last five minutes and the promise of that beer is helping.
RedheadinaMeadow_L004L.jpg

I hope you all enjoyed! Oh... and she got the beer!
 
So, is this VAM or DAZ being carefully painted over by SD?

It's only a matter of time before I make a proper "crux" model based on the images we have in our archives. It'll kick ass to have that extra layer of realism. Especially since Daz refuses to have good looking hair to my eyes, like ever. I would kill for some temporal consistency, personally; I'm making a cross model that's nearly that realistic right out of the render, so a low pass of 2-10% noise will go a lot further. Did you see the new photoshop tools? You are going to get me to talk shop about this, hahaha.
 
mp5stab said:
So, is this VAM or DAZ being carefully painted over by SD?

It's only a matter of time before I make a proper "crux" model based on the images we have in our archives. It'll kick ass to have that extra layer of realism. Especially since Daz refuses to have good looking hair to my eyes, like ever. I would kill for some temporal consistency, personally; I'm making a cross model that's nearly that realistic right out of the render, so a low pass of 2-10% noise will go a lot further. Did you see the new photoshop tools? You are going to get me to talk shop about this, hahaha.
Click to expand...
These images were made from some old images I had laying around, some from my photo shoot with Randy years ago, some were made with Poser. They were then loaded into SD and I blocked out the girl and said "a pretty redhead" along with a few other words of instruction. SD then makes a new girl to fill out the blanked out area. A hundred images later, I dug out these few and then spent some time fixing the hands, adding nails and blood in Gimp (I don't have Photoshop).

Oh, the first image was a mistake and is pure SD. SD can make beautiful photograph quality images of girls all day long. Just don"t try to put them on a cross, or make a mermaid, or a sexy girl robot. It falls apart after that. As for hands? Well.... It has huge problems with hands. It can get them perfect and it can also butcher them. Some images it comes up with have me busting a gut laughing.

You do know that you can download SD for free and make the images any size you want, don't you. You will need a pretty good Nvidia GPU to do it. I had a Radeon GPU that was just good enough, but SD does not work with SD. SO.... I bought a new GPU with 12mb of ram and it can crank out the base images every ten seconds.

I would love to see some of the images you make of this character that is almost photo realistic. That was always my goal when I was using Poser. I had to go to a different render engine because I never could get results I liked in Power. The engine no longer is supported as Microsoft disabled completely Internet Explorer and for some reason Reality needs that. So, it looks like I am done with Poser. I wish I could get going with DAZ, but I am still mad at them for abandoning Poser, I also had a TON of content for Poser. I just refuse to go back and start all over. I am getting old, hardheaded and grumpy!

MH
 
Oh, stable diffusion is already in my workflow. Those fireworks in the background of my new years post were 100% a stable diffusion output behind an alpha-mapped render. I feel your pain for poser, you were one of the early poser artists that inspired me to try my hand at blender. I think the move might be to do the hands in vam/poser/Daz and be very conservative with the demosing on imgtoimg. Finisher has been playing around with inpainted versions of his renders, that’s one approach.
 
mp5stab said:
Oh, stable diffusion is already in my workflow. Those fireworks in the background of my new years post were 100% a stable diffusion output behind an alpha-mapped render. I feel your pain for poser, you were one of the early poser artists that inspired me to try my hand at blender. I think the move might be to do the hands in vam/poser/Daz and be very conservative with the demosing on imgtoimg. Finisher has been playing around with inpainted versions of his renders, that’s one approach.
Click to expand...
I am honored to hear that I was one of those that inspired you!

I think you are right about using VAM and then SD. I also have Character Creator, which is pretty good. All of the Roxanne pics were done in it. I like Character Creator because I can make clones of real people pretty easy.

Yeah, I had to bring hands in from other sources on all of these images, as well as the feet. SD really does not like to do women on crosses. It will do men all day long, but always messes up women. Don't look too close at these images, they are not that well done. I was in a hurry to get them all done today.

MH
 
I think that goes for manips as well. Look closely enough and you'll always find some imperfection, but whatever you say about it, those are still fantastic images from where I'm sitting.
 
SD doesn’t do women on crosses because it doesn’t have references for that outside of artsy full clothed photo shoots. You gotta train a model, make an embedding. Then when you type the keyword like “femcrux” as an example it knows you are trying to create an image in the style of all of the images you fed it. Sometime soon I might make this model for the cruxforums; I can do crazy stuff like ask each artist individually if they want to be in the model or not. You know, because that’s 1. Ethical and 2. Because I can. It would be a lot of work. Labeling, sorting, and organizing the training data will take a long time, but by the end, it will “know” how to make a crucified woman if I input “femcrux”. Alternatively I could just make smaller embeddings for nailed feet and wrists, and start with the poses on Daz. It’s… a thought. But if one of us does that shitload of work + all of the fantastic crux art that has been made over the years, you could have your one-sentence model.
 
mp5stab said:
SD doesn’t do women on crosses because it doesn’t have references for that outside of artsy full clothed photo shoots. You gotta train a model, make an embedding. Then when you type the keyword like “femcrux” as an example it knows you are trying to create an image in the style of all of the images you fed it. Sometime soon I might make this model for the cruxforums; I can do crazy stuff like ask each artist individually if they want to be in the model or not. You know, because that’s 1. Ethical and 2. Because I can. It would be a lot of work. Labeling, sorting, and organizing the training data will take a long time, but by the end, it will “know” how to make a crucified woman if I input “femcrux”. Alternatively I could just make smaller embeddings for nailed feet and wrists, and start with the poses on Daz. It’s… a thought. But if one of us does that shitload of work + all of the fantastic crux art that has been made over the years, you could have your one-sentence model.
Click to expand...
I read an article about training the AI and building models that knows specific people, like all of your friends, but the process sounded to complex for me. Not just all the keypunch work, but all the computer code stuff. I imagine that eventually someone will build a windows program where you load all the images and it trains the AI. There is little doubt about two things; 1) we are in the earl days of this tech, and 2) there is no such thing as "photographic" or even "video evidence" anymore.

I think it would be great if you trained an AI in crux art. If you do, I will be your first customer!

MH
 
mermaidhunter said:
Well, it began to seem to me that nobody was going to post some redheads on crosses, so I threw some together.

A redhead bound being marched out to her crucifixion site.

I hope you all enjoyed! Oh... and she got the beer!
Click to expand...

Fantastic. Amazing.

mermaidhunter said:
These images were made from some old images I had laying around, some from my photo shoot with Randy years ago, some were made with Poser.
Click to expand...

Yes, the poses were unmistakable.

I would love to see more live crux from you MH, but this goes a long way towards meeting my crux need, so thank you.
Jollyrei said:
I think that goes for manips as well. Look closely enough and you'll always find some imperfection, but whatever you say about it, those are still fantastic images from where I'm sitting.
Click to expand...

Adopt the attitude of the medieval monkish illustrators, Jolly, the imperfections show that you are human, to be perfect would be to challenge God!

As for SD, before I get into downloading and playing with that I think I'd have to seriously upgrade my GPU!
 
phlebas said:
Fantastic. Amazing.



Yes, the poses were unmistakable.

I would love to see more live crux from you MH, but this goes a long way towards meeting my crux need, so thank you.


Adopt the attitude of the medieval monkish illustrators, Jolly, the imperfections show that you are human, to be perfect would be to challenge God!

As for SD, before I get into downloading and playing with that I think I'd have to seriously upgrade my GPU!
Click to expand...
Thank you Phlebas, I appreciate your comments!

As for installing SD and having a go at it, I thought I would point out that you can also do CPU processing, but is is VERY slow. I tried it that way at first and once I realized it was something I wanted to dive into, I broke down and bought a new GPU. But, you can get your feet wet with cpu processing.. I think with CPU processing it was something like over a minute per image for a 512 X 512 image. With the new GPU installed I can crank out about 10 of those images per minute. In the end, at least for me, it is a numbers game. Since I am always trying to do things that are not in it's database, such as crux, female robots, and mermaids the AI has a hard time. SO, I have to pump out a lot of images to get one that is useful.

MH
 
mermaidhunter said:
For anyone interested, here is a link to my Deviantart page. There are no crux images there, only robots and mermaids. I am working on a collection of Harem Girls, but it is not ready for daylight yet.


MH
Click to expand...
I love all of your art, but I deeply love your VAM work. With the black girl especially; I find myself returning to that one very often. Maybe you could try to crucify a redhead too?
 
mp5stab said:
I love all of your art, but I deeply love your VAM work. With the black girl especially; I find myself returning to that one very often. Maybe you could try to crucify a redhead too?
Click to expand...
Than you mp5stab, I appreciate that. Especially from an artist of your caliber.

I really like the black girl in the video I made as well, the sad part is that her beauty did not really shine in that lighting. Sometimes I get stuck in an idea and how I want things to be. So, I work and fight and try to get them the way I want, and when it doesn't work out, I either walk away from that project or accept a compromise. For some reason, I never consider changing direction to get a better result from my original intent. I always want that original vision I had in my head.

I really should make another with her as well, in different lighting. She is strikingly beautiful in my eyes.

MH
 
Now hang on a minute there, @mermaidhunter you watched me become an artist of your caliber. You were here back when poser was the latest thing, and I was still doing pencil sketches in my school notebooks. And now, here we are, with realtime raytracing, AI art, and VR crucifixions. It's a bright future for artists of our caliber. We just have to keep up with it!

(Also, I'm back to writing stories, if you'd like to check that out)

Doing that crux in different lighting might be fun, but have you tried having a girl wear a ponytail in VAM? If it has good hair physics for ponytails, I might lose a sizeable portion of my life making VAM animations very soon.
 
mp5stab said:
Now hang on a minute there, @mermaidhunter you watched me become an artist of your caliber. You were here back when poser was the latest thing, and I was still doing pencil sketches in my school notebooks. And now, here we are, with realtime raytracing, AI art, and VR crucifixions. It's a bright future for artists of our caliber. We just have to keep up with it!

(Also, I'm back to writing stories, if you'd like to check that out)

Doing that crux in different lighting might be fun, but have you tried having a girl wear a ponytail in VAM? If it has good hair physics for ponytails, I might lose a sizeable portion of my life making VAM animations very soon.
Click to expand...
The fact that I started before you doesn't take anything away from your accomplishments and abilities now. So, when you compliment my work, it means a lot. Besides you are apparently more gifted as it took me a lot longer to get where I am and you have or will soon surpass me. Nothing negative or competitive meant by that, just recognizing your talent. For me the fun is in the making of the art, sharing it is just icing on the cake.

MH
 
Yeah... but like I'm not just "going to surpass you", I love making 3d models, sculpts, I have even tried my hand at environments recently. And I've shared a lot of that stuff with other 3d artists. I like to think a rising tide lifts all boats. If VAM can take genesis models, just tell me what type because you might be about to get some custom figures from me that I sculpted a few years to a few months ago. For me. the sharing is part of the art process. Finished renders are one thing, but I can do so much more than that. how many times has my cross popped up in a random render? That's my art. :D Or at least, part of it. I have a discord, by the way, if you want to connect/share/join the secret cabal of artists I've assembled.
 
I think the current version uses Genesis 2 models, the long promised but yet to be delivered Version 2.0 of VAM will use Genesis 8 models. I don't do any importing or exporting of models because Poser is not compatible and I am too hardheaded to learn DAZ. So, that is about all I know about that aspect other than you will need some plugins to facilitate the exchange.

By the way I was in VR with VAm last night and noticed that the girl had a ponytail. It moved around like a normal ponytail. I will record some video when I get a chance and post it.

MH
 
You must log in or register to reply here.
Back
Top Bottom