Posted on my pixiv page: https://www.pixiv.net/en/artworks/131691237
As today is 18th June, the day of Hatate, I Illustrated Hatate Himekaidou with her remarkable quote "It's a giant meddling!" at Obaniwan Channel, a popular MMD movie channel of Touhou fan-art.
Posted on my pixiv page: https://www.pixiv.net/en/artworks/131599386
One scene of Marisa and Alice dispatching in Touhou 10 Imperishable Night shown in an anime Touhou, The Memories Of Phantasm.
Their facial expressions here also reflect the true nature of their relationship — a rivalry marked by playful teasing, even as they form an occasional alliance when their interests align.
Illustrated a pair of Nue Houjuu and Sekibanki hugging each other as per request.
Posted on pixiv: https://www.pixiv.net/en/artworks/131387646
* AI hybrid experiment*
Asked ChatGPT to redraw some unfortunate illustration with critical mistakes.
A set of these illustrations is a redrawn version by ChatGPT, of the following works
https://www.pixiv.net/en/artworks/120485457
https://www.pixiv.net/en/artworks/119309631
Used in these short films:
My feedback on AI art (both pure and hybrid):
This is actually my first attempt at creating AI art. However, rather than instructing an AI tool to generate artwork via a command prompt—such as using Python in Stable Diffusion—I used an illustration I had already created and had the AI—ChatGPT, in this case—process it.
It is an interesting experiment to upload my already made images to an AI tool such as ChatGPT to redraw it. The AI corrects the crucial mistakes I have made while transforming to a popular art style. It can be useful for cases like
The drawback is losing my style of characteristics in these arts after the AI process. While observing various AI arts, the art style of these AI tools seems to be quite unified to the certain extend. AI is still generally limited to reproducing art styles that are well represented in its training data such as historically famous or currently popular arts. Because a unique art style of the user is usually undermined, it is not suitable for creating one piece of their impressive art.
Furthermore, only a very limited number of files are available process with these AI tools for a free option. It can be far less efficient to continue processing an image with AI in order to achieve the desired result than simply using traditional hand-drawing techniques.
In conclusion, AI art in any form has pros and cons. It depends on its application and the characteristics of the user's art skill. I may use the AI in some situation such as when I need a mass production of images. On the other hand, I am confident in my own drawing technique even though it might not perfectly match with what the popularity requires at the moment. Moreover, I am passionate to keep practising and improving my art technique. Hence, the AI process is useful as a supplement; not a substitution.
The films with these AI-hybrid except for the last GIF (my original)
Posted on my pixiv page: https://www.pixiv.net/en/artworks/131124657
With voice
YouTube: https://www.youtube.com/shorts/mg6l3roR-94
NicoNico: https://www.nicovideo.jp/watch/sm45049841
I have created the standing picture of Nareko Michigami for my own Yukkuri kaisetsu!
Posted on my pixiv page: https://www.pixiv.net/en/artworks/130355468
My hybrid of analogue for drawing shapes and digital for painting colour
My admirable Adam Smith watching out the world from Royal Mile, Edinburgh, Scotland, UK
Posted on my pixiv: https://www.pixiv.net/en/artworks/129895830
The free Gensokyo alliance of Takane Yamashiro and Byakuren Hijiri in my fan-art story
Used in my YouTube:
Posted on my pixiv page: https://www.pixiv.net/en/artworks/129743439
自家用ゆっくり解説用の自作山城たかねの立ち絵を作りました！
I have created the standing picture of Takane Yamashiro for my own Yukkuri kaisetsu!
