[AI Image Generation] Small Theater: Fluffy Family 😸🐶 + Deep Dive into Model Selection: SeaArt.AI🐳

Thank you for finding this article🥰

I'm Natsume, and I want to walk alongside AI while using the AI that evolves every day😊

When it comes to AI image generation, I have my own personal 'composition' for it🥰

That is

Dogs, cats, and women.

Mostly cats... though lol🤣

I keep creating them because I love them and can keep doing it. This is something I started with nijijourney, but since the journey series holds the potential for various expressions, I decided to keep it on the bench and create with a different AI instead🥰

I wonder what kind of depictions will happen if I create various cats with various AI tools?

It's a very good adventure🥰

However, there are also many 'What on earth is this!!' moments due to AI 'quirks'🥰
This is what makes it interesting🥰

This time, I would like to do my best with SeaArt, which I have been using recently.

Well then, with me, Natsume, and the black cat, let's go meet the fluffy ones🥰

*Entering a small theater, I would appreciate it if you could join me.

Well, hello there! 🥰

To think I'd meet the twins and a nice, chubby Shiba (my absolute favorite)! ✨

This plump Shiba Inu is just my type... honestly... 🤤 And I realized that on YouTube, haha! 🤣

Now, let's move on.

Oh my! Another Shiba! (Natsume! How much do you love Shiba Inus?!)

So big!

The animals in this model are just "huge!"

Maybe that's why I like them, but the plumpness is incredible!!

Well, well! 🥰

Is that a retriever type? They look like they're having fun, which is great! 🥰

It's probably best to leave them be. 💦

Must be nice... having a relationship like that... (stares) (　＾ω＾)... (Jealous of AI image generation, lol 🤣)

Uh... uh..., I-I see, that is wonderful.💓
(Let's just leave them be.)

I see, I see. They're sitting at the table watching you, aren't they?

I see, so that's what was going on.🥰

Hm?

~~♬♬♬
Oh, I'm getting a call.

Natsume: "Huh? You want me to come over now?"

Sorry, my sister suddenly told me to come over, so I'm going to her place.💦

What a defenseless sister.💦

Natsume: "Sis!! Did you fall back asleep!? Seriously~ with Mike too!? You lucky person!! That's so unfair! (´;ω;｀)"

Whoa!?

Natsume: "Reika-san, my sister's friend!! And her dog Lucky! What are you doing here without the lights on!?💦"

It seems they had plans together today, and although they were let into the house, my sister didn't wake up, so they were just waiting.

You know, you're allowed to turn on the lights, right??💦

Phew, I guess it's time to head home(●´ω｀●)

Oh! Isn't that Michiko-chan!!

Ah, well, I have plans to hang out with Michiko-san, so!!

See you later~~!!

Don't worry! I'm wearing them!!🤣

So, how was it?

The occasional Natsume Short Theater lol

I hope you found it interesting, healing, or otherwise enjoyable🥰

And now, a slightly serious topic to finish off.

Are you thinking about trying out AI tools that use "models"?

For those who are thinking that, and of course for those who aren't,

I, Natsume, would like to share my thoughts on

how to approach "models"🥰
Please keep in mind that this is just my personal opinion, so feel free to just take it with a grain of salt lol

Is SeaArt.AI an AI tool for choosing "models"? It goes without saying that the "criteria" for selection depend on the individual, but this is surprisingly difficult.

In this article, I am using three types of models.
Can you tell the difference because the touch is different?

Actually, to get to this point, I have posted the completed illustrations on note, taking into account the quirks and drawing styles of each model.

Until the desired object reaches "completion".

That path is not easy.

When choosing a model, I look at the "character" in the thumbnail to decide whether to create it🥰

And, when creating, several patterns emerge.

① I like the character ⇒ I can't create it well, there are limits to the range (composition) I can create💦
(Give up)

② I like the character ⇒ I created it well, there are limits to the range (composition) I can create💦
(Continue)

③ I like the character ⇒ I can't create it well, there seems to be a wide range (composition) I can create✨
(Give up)

④ I like the character ⇒ I created it well, there seems to be a wide range (composition) I can create✨
(Continue)

With these patterns, which one do I choose to create with?
The ideal is ④, but encountering pattern ④ is truly a blessing🥰

Here, the point is the "seems to be" in "seems to be a wide range (composition) I can create".
The reason I can't say for sure that it is wide is that as I create, I realize the limitations, just like in ②.

Of course, I think there is also a lack of skill.
But at this stage, perhaps due to my own immaturity, the result becomes "I can't do it".

So, it ends up being ②(●´ω｀●)

Conversely, let's think about ②.

Are you satisfied just looking at your own illustrations?

Beyond the thumbnail, when you look at the details of a "model," you can see various illustrations created by many different people.

Wait, you can draw illustrations like this with this model? How?

That's right 💦

You are placing "limitations" (satisfaction) on yourself. In reality, there are actually configurations that can draw even more.

So, if you dig deeper into (2), it leads to (4).

If it goes well, explore with that "model." The problem is that if you "explore," the AI might learn it, and there is a possibility you won't be able to go back. Therefore, it is best to keep a record of successful prompts so you can have them generated quickly ✨

And if you get bored of using the same model, go look at other models to find another partner for yourself.

By doing so, you can expand the range of your illustrations.

I think this is also the real thrill of AI image generation 🥰

How about it?
Does this seem easy?

...It actually takes patience and effort, you know? 🥰

It's just one illustration.
Only the person themselves knows what kind of background that illustration has.

With what thoughts?
In what way?
What kind of self?

"Expressing" things is truly not an easy task, is it?

Besides, AI image generation has increased, and we are now flooded with wonderful illustrations.

Things like, "So beautiful," "So cute."

Because of that, our eyes have become quite accustomed to beautiful and cute things.

"Originality," "How to present it"

I think we are living in an era where that is highly sought after.

I think this article is the type that just ends with "It's pretty" or "It's cute"💦

Originality and how you present your work mean making the viewer feel that there is something else hidden in the illustration, another meaning or intention beyond just being "pretty."
I feel that we must generate at that level using AI.

AI image generation has a high barrier to entry.

It happens often, doesn't it?
A beautiful girl you can draw with just "1girl."

This is true.

Especially if you choose the right "model," you can create a beautiful woman.
The problem is how to present her after you've chosen.

Ah... I'm falling into the rabbit hole...

Searching for your own "expression" and "depiction" is like searching for another version of yourself.
In other words, it means continuing to face AI head-on.

Apart from AI tools that don't involve models, there is this SeaArt.AI🐳 for model selection.

I'm deep in the rabbit hole, so deep.

What if, I think this way, or maybe it's like this?

Like, "This is how I usually think when I generate images!"

I would be very happy if we could exchange opinions on image generation and such🥰

If you don't mind, I would be happy to receive your opinions and impressions.

This ended up being long, didn't it💦
Thank you for sticking with me until the end✨

I am always so happy to receive your likes, comments, and magazine subscriptions🥰
Thank you so much (´;ω;｀)

Well then, have a great AI image generation life~💐

Behold!!

This is heaven!!

So huge lol🤣