AI AP sequences

[BBB]

If you want an image sequence then in theory, telling an AI image generator to make a multi-panel image is a good idea. Then (assuming it does enough pixels to fit a whole sequence) it can give the subject, background, art style etc. a consistency that it frankly can't when generating each panel separately.

A problem I had was, the stable diffusion generator I was using couldn't understand relational directions. You could tell it "on the LEFT is a short woman and on the RIGHT is a tall woman" all you want, but it wouldn't understand.

This week, Perchance's AI text-to-image generator got an upgrade, and now it understands exactly this kind of instruction. So I gave it a go. Here are my best results trying to make a 3-panel age progression sequence.

https://www.imagebam.com/view/GAEWXV

I feel like with AI, the barriers to content creation are coming down. If anyone's so inclined, they can enter a prompt like I did and churn out as much free content as we see on this forum in a month!

I did feel bad that the prompt I settled on included "by a top rated Western fetish artist on DeviantArt." However for what it's worth, I make a point of not asking the AI for any particular artist's style by name.

[lefix]

Greetings for you share. Really good transitions. Can you share the promts tha t you use , cause i tried on Perchance's AI but i'm having issues on instructing AI in the way i imagine the sequence.

[BBB]

Originally Posted by lefix

Greetings for you share. Really good transitions. Can you share the promts tha t you use , cause i tried on Perchance's AI but i'm having issues on instructing AI in the way i imagine the sequence.

Thanks!

So prompt writing is different than before - see https://perchance.org/new-image-model-notes. Now you can, and apparently should write more natural sentences instead of a list of tags. Finding multiple ways to express the same idea helps. As for my prompt, I am constantly changing it to see what works better or worse through trial and error. Here is a more general purpose form of what I've used, which was yielding good results (keep in mind "good results" is still less than 50% good ones).

***

Art Style: No Style

Description:
3-panel age progression time-lapse growth sequence of an attractive female growing from her early teens to late teens. It shows her in sequence at each year at age 13, 15, and 19. It resembles a police lineup and is inspired by educational diagrams of the Tanner stages of puberty. Her transformation progresses linearly from left to right. With each year she is bigger, hairier, and more mature than before. Her sexual characteristics are labeled and they wobble and quiver dynamically with motion lines.
In the middle is a burgeoning 5'3" 15 year old adolescent in the middle of puberty at Tanner stage 3 with B-cup breasts and an intermediate amount of body hair.
To her left is a cute 5'1" and completely smooth 13 year old at Tanner stage 2 with little body hair and AA-cup breast buds.
To her right is a smugly satisfied more sexually mature 5'5" 19 year old at Tanner stage 5 with a more shapely feminine body, C-cup breasts, and more pubic hair, armpit hair, and leg hair.
It's a full body shot. It's a dramatic breathtaking best quality masterpiece with hyperrealistic anatomically correct proportions and intricate detail and composition by a top rated fetish artist on DeviantArt.

Anti-Description:
penis, intersex, futa, pregnant

***

If you use some or all of this, I encourage you to play around with it and make it your own. Once you're getting the type of image you want, you can start adding or adjusting details. Add a background scene, change her hair or skin color, change her reaction, switch from AP to a different transformation entirely... whatever you like!

[lefix]

Really thenk's , your advice helped me a lot.
I played something around with different filters and syntax.
I need to improve a little paragraph and lists prompts , but i'm enjoing in the process.
In the mean time these are the best results :

1 set of sequence - painted anime , drawn anime
https://www.anonfile.la/7c5e0c

2 " " " " " " - cinematic , casual photo , b-horror movie , wide angle shot
https://www.anonfile.la/64c90e

[godleydemon]

Interestingly enough, I've been using perchance for quite a while. I have a pretty extensive list of key words I normally use to get good generations. But with the current change, you have to be more descriptive. Which has posed it's own problems but opened interesting doors to generation.

This was my most recent attempts prompt

A 3 panel comic of a beautiful naked woman with generous curves looking younger, starting from left to right. She has blue eyes and has a stoic, cold demeanor look. She has incredibly long white hair that gets shorter from left to right. Her breasts are large and beautiful on the left, but shrink to small nearly flat when it get to the right. She is tall and leggy on the left and short, skinny and petite on the right. She is confident and sexy on the left and tearfully crying on the right. On the left is a mature beauty, on the far right the same woman is tiny and skinny. The middle image should always be the halfway point between the left and right. Full body, Best Quality, NSFW

I think what's more unique about my approach is probably my Negatives.

worst quality, bad quality, low quality, lowres, anatomical nonsense, artistic error, bad anatomy, interlocked fingers, extra fingers, text, artist name, signature, bad feet, extra toes, ugly, poorly drawn, censor, blurry, watermark, old, oldest, glitch, deformed, mutated, disfigured, long body, bad hands, missing fingers, extra digit, fewer digits, cropped, very displeasing, sketch, jpeg artifacts, username, censored, bar_censor, mosaic_censor, conjoined, bad ai-generated, long neck, skin blemishes, skin spots, acne, the wrong limb, error, black line, excess hands, elf, text

As you can see, I have quite a few negatives but it produces some pretty good results. I included a few results I've gotten with this setup to create sudo AR. Let me make this clear, I do not believe this goes against any of our rules here. As the woman in question is still of age, but smaller, skinnier and petite at the end

I'm unsure if indicating height, breast size, ect. will help much in generation if I'm being honest. The generator seems to like word descriptions like small, medium, large, extra large, huge, gigantic, humongous, gargantuan, ect. I also like using mature, curvy, petite, tiny and skinny. The generator just seems to understand this more instead of specific measurements. I've also found simply putting, "the middle image will always be the half way point between the left and right" seems to create pretty decent transitions without having to be specific. The generator just kind of figures it out. When I get off work tonight I'll play around with including some of OP's descriptions. For example, I get pretty interesting results when I include "It's a dramatic breathtaking best quality masterpiece with hyper-realistic anatomically correct proportions and intricate detail and composition by a top rated fetish artist on DeviantArt." that seems to up the quality of the generation. The last image in my attachments is with that sentence included.