# aiartwork developer fantasyartdiffusion development midjourney difnesfusion nes aiart watercolordiffusion ai saharandust atlanticocean aerosolopticaldepth retrogames retrogaming

have been busy today...
but few new dataset sneakpeaks

retro is still early stage. still training

Working on 4 datasets..

VintageDataset 256
nameless 512
retro 512
fantasy 256.

8 17
2022-07-29

updated its deal and the jump is significant. Details are cleaner, anatomy is less horrific, and concepts come through more clearly. I don't think its the full LIAON-3B dataset yet, so there's more to look forward to. But color me impressed.

0 2
2022-07-26

17 hours of training my transformer implementation on anime face dataset and i will train it more

0 19
2022-07-25

oh boy, that’s a complex one! You got the coder who figured out how to make silicon have dreams, but they used enormous datasets from numerous artists in order to train this AI model… Then we come in as artists and use that construct to come up with something original! complex!!

0 1
2022-07-20

1.0 -> 1.35 Release will be TODAY!!
Whole new datasets Style

Is your soul ready?
Back in time retrogaming

10 34
2022-07-19

Recognize this Artwork? I can train using your Artwork as Dataset.

“Girl with tiger face in the mountain”

So.. can i sell it? Thats the kind of question this article provide. 😉

0 1
2022-07-10

Taking hundreds of these images, I created a dataset to train a STYLEGAN2 model on in hopes to create "new" versions of my existing work.

Instead, I found that the stills and videos created were more beautiful than anything I'd done before.

0 0
2022-07-08

Trying out a different dataset~

5 27
2022-07-04

An early output from because I don't know how or when to stop. Dataset is half one I made a couple years ago, and half recently-gathered :)

1 20
2022-07-02

My custom finetuned model "Liminal Diffusion" beta version should be released tomorrow.

Created with my own hand curated and custom dataset this model has been trained to create eerie, strange, and surreal liminal spaces.

Very happy with how it has turned out! Free4All.

1 2
2022-06-24

Some good games there. 👍
My old breadbin setup looked very similar. (Same basic model of Datasette, except it was the standard cream coloured one.) And, of course, the joysticks changed over the years, as each one inevitably broke.
😋

1 3
2022-06-22

Ongoing major transport episode across This analysis shows the path of thick plume since 1 June.
Animation created using CLEOPE, the ONDA streamlined access service to CAMS datasets.
➡️ https://t.co/LAoBBuwGNU

8 20
2022-06-16

Further progress, fixed a bug in correcting face orientation when building the training dataset, which ought to improve average quality further

0 0
2022-05-26

Anime GAN project continues, with a more carefully curated dataset and various architectural adjustments

0 1
2022-05-26

Samhain judges the spirits that come to his realm, deciding what next stage of the afterlife will be for each one.

The Lord of the Dead - King's Quest 6. Upscaled with ESRGAN w/Manga109 dataset model by MusicallyInspired https://t.co/x0tE2qqe6M

3 35
2022-05-21

The art of creating a dataset (for StyleGAN interpolation animation)
It's all about finding the balance between diversity and similarity. Too diverse, and the training process 'collapses'. Too similar, and the results are uninteresting. Currently making these with VQGAN/CLIP.

5 33
2022-05-21

Add residual to latent diffusion first stage training and train with danbooru dataset. Now it can encode 256x256x3 pictures to eight ordered 16x16x4 latent codes. I don't know if this first stage model generated latent codes can be used in diffusion. https://t.co/5NFcO64qDK

0 6
2022-05-20

It takes multiple layers of custom dataset training, typically over weeks, to arrive at a model that will simulate/synthesize exactly what I want. In an era of commercial click-to-generate AI art apps, I try to make sure the aesthetic qualities of my work help it to stand out /19

0 22
2022-05-19